├── .flake8 ├── .github ├── ISSUE_TEMPLATE │ ├── bug_report.md │ ├── custom.md │ └── feature_request.md ├── PULL_REQUEST_TEMPLATE.md └── workflows │ ├── docker.yaml │ ├── docker │ └── docker-compose.yaml │ ├── pre-commit.yaml │ ├── release.yaml │ ├── sphinx-doc.yaml │ └── unittest.yaml ├── .gitignore ├── .pre-commit-config.yaml ├── CONTRIBUTING.md ├── LICENSE ├── README.md ├── README_zh.md ├── benchmark ├── README.md ├── bench.py ├── config │ ├── alfworld-template.yaml │ ├── countdown-template.yaml │ ├── frozenlake-template.yaml │ ├── gsm8k-template.yaml │ └── guru_math-template.yaml ├── plugins │ └── guru_math │ │ ├── naive_dapo.py │ │ └── reward.py ├── reports │ ├── alfworld.md │ └── frozenlake.md └── scripts │ ├── gen_countdown_data.py │ ├── gen_guru_math_data.py │ └── get_alfworld_full_data.py ├── docs ├── README.md └── sphinx_doc │ ├── Makefile │ ├── _static │ └── logo.svg │ ├── _templates │ ├── article_header_customized.html │ ├── language-switch.html │ ├── package.rst_t │ └── versions.html │ ├── assets │ ├── DYN-NCCL.png │ ├── DYN-STATEDICT.png │ ├── FIXED-NCCL.png │ ├── FIXED-STATEDICT.png │ ├── RAFT_alfworld_reward_curve.png │ ├── agentscope_dapo_qwen3-4B_reward.png │ ├── agentscope_dapo_reward.png │ ├── agentscope_dapo_turns.png │ ├── agentscope_gsm8k_reward.png │ ├── agentscope_gsm8k_turns.png │ ├── agentscope_websearch_reward.png │ ├── agentscope_websearch_turns.png │ ├── alfworld_reward_curve.png │ ├── alfworldv2_reward.png │ ├── async-curve.png │ ├── bench_alfworld_step.png │ ├── bench_alfworld_time.png │ ├── bench_frozenlake_step.png │ ├── config-manager.png │ ├── countdown-bench.png │ ├── data-in-labelling-page.png │ ├── data-in-projects.png │ ├── data-labelling.png │ ├── data-projects.png │ ├── email_eval_accuracy.png │ ├── email_reward_mean.png │ ├── email_rollout_accuracy.png │ ├── example_experience_replay │ │ ├── exp_replay_X_explore_step.png │ │ ├── exp_replay_X_time.png │ │ └── exp_replay_model_version.png │ ├── geometry3k_qwen25_vl_3b_reward.png │ ├── grpo_rubric_reward.png │ ├── gsm8k-bench.png │ ├── gsm8k_ruler_eval_accuracy.png │ ├── gsm8k_ruler_gold_reward.png │ ├── gsm8k_ruler_judge_success.png │ ├── gsm8k_ruler_reward.png │ ├── gsm8k_trainable_ruler_eval_accuracy.png │ ├── gsm8k_trainable_ruler_gold_reward.png │ ├── gsm8k_trainable_ruler_judge_success.png │ ├── gsm8k_trainable_ruler_reward.png │ ├── gsm8k_trainable_ruler_reward_for_judger.png │ ├── guru-bench.png │ ├── mix_vlm_reward.png │ ├── opmd-curve.png │ ├── scripts-multi-plot.png │ ├── toolace_3b_response.png │ ├── toolace_3b_rewards.png │ ├── toolace_length_curve.png │ ├── toolace_reward_curve.png │ ├── trinity-architecture.png │ ├── trinity-design.png │ ├── trinity-mix.png │ ├── trinity-mode.png │ ├── trinity-multi-step-compare.png │ ├── trinity-title.png │ ├── trinity_data_process.png │ └── webshop_reward_curve.png │ ├── build_doc.sh │ ├── make.bat │ ├── source │ ├── api_reference.rst │ ├── conf.py │ ├── index.rst │ ├── main.md │ └── tutorial │ │ ├── develop_algorithm.md │ │ ├── develop_operator.md │ │ ├── develop_overview.md │ │ ├── develop_selector.md │ │ ├── develop_workflow.md │ │ ├── example_async_mode.md │ │ ├── example_data_functionalities.md │ │ ├── example_dpo.md │ │ ├── example_megatron.md │ │ ├── example_mix_algo.md │ │ ├── example_multi_turn.md │ │ ├── example_react.md │ │ ├── example_reasoning_advanced.md │ │ ├── example_reasoning_basic.md │ │ ├── example_search_email.md │ │ ├── example_step_wise.md │ │ ├── faq.md │ │ ├── synchronizer.md │ │ ├── trinity_configs.md │ │ └── trinity_installation.md │ └── source_zh │ ├── api_reference.rst │ ├── conf.py │ ├── index.rst │ ├── main.md │ └── tutorial │ ├── develop_algorithm.md │ ├── develop_operator.md │ ├── develop_overview.md │ ├── develop_selector.md │ ├── develop_workflow.md │ ├── example_async_mode.md │ ├── example_data_functionalities.md │ ├── example_dpo.md │ ├── example_megatron.md │ ├── example_mix_algo.md │ ├── example_multi_turn.md │ ├── example_react.md │ ├── example_reasoning_advanced.md │ ├── example_reasoning_basic.md │ ├── example_search_email.md │ ├── example_step_wise.md │ ├── faq.md │ ├── synchronizer.md │ ├── trinity_configs.md │ └── trinity_installation.md ├── environments ├── data.yaml ├── env_mapping.json └── training.yaml ├── examples ├── RAFT_alfworld │ ├── RAFT_alfworld_7B.yaml │ ├── RAFT_reflect_alfworld_7B.yaml │ └── README.md ├── agentscope_react │ ├── README.md │ └── gsm8k.yaml ├── agentscope_tool_react │ ├── README.md │ ├── agentscopev0_tool_react_dapo.yaml │ ├── agentscopev0_tool_react_gsm8k.yaml │ └── agentscopev1_tool_react_dapo.yaml ├── agentscope_websearch │ ├── README.md │ ├── agentscopev1_websearch_agent.yaml │ └── get_webwalkerQA_data.py ├── asymre_gsm8k │ ├── README.md │ └── gsm8k.yaml ├── asymre_math │ ├── README.md │ └── math.yaml ├── async_gsm8k │ ├── README.md │ ├── explorer.yaml │ ├── run.sh │ └── trainer.yaml ├── bots │ ├── README.md │ ├── README_zh.md │ ├── bots.yaml │ ├── random.yaml │ └── workflow │ │ ├── bots_math_boxed_reward.py │ │ ├── bots_math_boxed_workflow.py │ │ └── bots_reward.py ├── cispo_gsm8k │ ├── README.md │ └── gsm8k.yaml ├── dapo_math │ ├── README.md │ └── dapo.yaml ├── dpo_human_in_the_loop │ ├── README.md │ ├── demo-data.jsonl │ └── dpo.yaml ├── dpo_humanlike │ ├── README.md │ └── dpo.yaml ├── grpo_alfworld │ ├── README.md │ ├── alfworld.yaml │ └── get_alfworld_data.py ├── grpo_alfworld_general_multi_step │ ├── README.md │ └── alfworld.yaml ├── grpo_email_search │ ├── README.md │ └── email_search.yaml ├── grpo_frozen_lake │ ├── README.md │ ├── frozen_lake.yaml │ ├── frozen_lake_reward.png │ ├── frozen_lake_test_score.png │ └── get_frozen_lake_data.py ├── grpo_gsm8k │ ├── README.md │ └── gsm8k.yaml ├── grpo_gsm8k_experience_pipeline │ ├── README.md │ ├── dj_scoring_exp.yaml │ └── gsm8k.yaml ├── grpo_gsm8k_ruler │ ├── README.md │ └── gsm8k_ruler.yaml ├── grpo_gsm8k_task_pipeline │ ├── README.md │ └── gsm8k.yaml ├── grpo_gsm8k_trainable_ruler │ ├── README.md │ └── gsm8k_ruler.yaml ├── grpo_lora_gsm8k │ ├── README.md │ └── gsm8k.yaml ├── grpo_math │ ├── README.md │ └── math.yaml ├── grpo_rubric_as_reward │ ├── README.md │ └── rubric.yaml ├── grpo_sciworld │ ├── README.md │ ├── get_sciworld_data.py │ └── sciworld.yaml ├── grpo_toolcall │ ├── README.md │ ├── get_toolace_data.py │ └── toolace.yaml ├── grpo_vlm │ ├── README.md │ └── vlm.yaml ├── grpo_webshop │ ├── README.md │ ├── get_webshop_data.py │ └── webshop.yaml ├── learn_to_ask │ ├── README.md │ ├── data_prepare │ │ ├── 1_info_extract_pipeline.py │ │ ├── 2_build_dataset.py │ │ ├── 3_rollout_then_evaluate.py │ │ ├── llm_info_extraction.py │ │ └── message_splitter.py │ ├── train.yaml │ └── workflow │ │ ├── prompt_learn2ask.py │ │ └── workflow_learn2ask.py ├── mix_chord │ ├── README.md │ ├── eval_bfcl │ │ ├── bfcl_reason_support.patch │ │ └── view_single_model_result.py │ ├── get_openr1_data.py │ ├── mix_chord.yaml │ └── mix_chord_toolace.yaml ├── mix_math │ ├── README.md │ └── mix_math.yaml ├── mix_vlm │ ├── README.md │ └── mix_vlm.yaml ├── opmd_gsm8k │ ├── README.md │ └── opmd_gsm8k.yaml ├── ppo_countdown │ ├── README.md │ └── countdown.yaml ├── ppo_countdown_exp_replay │ ├── README.md │ └── countdown.yaml ├── ppo_countdown_megatron │ ├── README.md │ └── countdown.yaml ├── rec_gsm8k │ ├── README.md │ └── gsm8k.yaml ├── sft_mot │ ├── README.md │ └── sft.yaml ├── sppo_gsm8k │ ├── README.md │ └── gsm8k.yaml └── topr_gsm8k │ ├── README.md │ └── gsm8k.yaml ├── pyproject.toml ├── scripts ├── docker │ ├── Dockerfile │ ├── Dockerfile.megatron │ └── Dockerfile.uv ├── install.py ├── multi_exps_plot │ ├── README.md │ ├── multi_exps_plot.py │ └── plot_configs.yaml └── start_servers.py ├── setup.py ├── tests ├── __init__.py ├── algorithm │ ├── __init__.py │ ├── advantage_fn_test.py │ ├── kl_fn_test.py │ └── policy_loss_test.py ├── buffer │ ├── __init__.py │ ├── experience_pipeline_test.py │ ├── experience_storage_test.py │ ├── file_test.py │ ├── formatter_test.py │ ├── queue_test.py │ ├── reader_test.py │ ├── reward_shaping_mapper_test.py │ ├── sql_test.py │ ├── task_scheduler_test.py │ └── task_storage_test.py ├── cli │ ├── __init__.py │ └── launcher_test.py ├── common │ ├── __init__.py │ ├── config_test.py │ ├── experience_test.py │ └── vllm_test.py ├── explorer │ ├── __init__.py │ ├── explorer_test.py │ ├── scheduler_test.py │ ├── step_wise_workflow_test.py │ └── workflow_test.py ├── manager │ ├── __init__.py │ ├── data │ │ └── train.jsonl │ └── synchronizer_test.py ├── service │ ├── __init__.py │ └── data_juicer_test.py ├── template │ ├── config.yaml │ ├── custom_chat_template.j2 │ └── data │ │ ├── countdown │ │ ├── test.jsonl │ │ └── train.jsonl │ │ ├── eval_long │ │ └── test.jsonl │ │ ├── eval_short │ │ └── test.jsonl │ │ ├── geometry │ │ └── train.parquet │ │ ├── gsm8k │ │ └── train.jsonl │ │ ├── human_like │ │ └── train.jsonl │ │ ├── sft_for_gsm8k │ │ └── sft.jsonl │ │ ├── sft_with_tools │ │ └── sft_with_tools.jsonl │ │ └── task_scheduler │ │ ├── subset_1 │ │ └── train.jsonl │ │ └── subset_2 │ │ └── train.jsonl ├── tools.py ├── trainer │ ├── __init__.py │ └── trainer_test.py └── utils │ ├── __init__.py │ ├── eval_utils_test.py │ ├── log_test.py │ ├── plugin_test.py │ ├── plugins │ ├── __init__.py │ ├── dependencies.py │ ├── main.py │ ├── my_monitor.py │ └── my_workflow.py │ └── registry_test.py └── trinity ├── __init__.py ├── algorithm ├── __init__.py ├── advantage_fn │ ├── __init__.py │ ├── advantage_fn.py │ ├── asymre_advantage.py │ ├── grpo_advantage.py │ ├── multi_step_grpo_advantage.py │ ├── opmd_advantage.py │ ├── ppo_advantage.py │ ├── rec_advantage.py │ ├── reinforce_advantage.py │ ├── reinforce_plus_plus_advantage.py │ ├── remax_advantage.py │ └── rloo_advantage.py ├── algorithm.py ├── entropy_loss_fn │ ├── __init__.py │ └── entropy_loss_fn.py ├── key_mapper.py ├── kl_fn │ ├── __init__.py │ └── kl_fn.py ├── policy_loss_fn │ ├── __init__.py │ ├── chord_policy_loss.py │ ├── cispo_policy_loss.py │ ├── dpo_loss.py │ ├── gspo_policy_loss.py │ ├── mix_policy_loss.py │ ├── opmd_policy_loss.py │ ├── policy_loss_fn.py │ ├── ppo_policy_loss.py │ ├── rec_policy_loss.py │ ├── sapo_policy_loss.py │ ├── sft_loss.py │ ├── sppo_loss_fn.py │ └── topr_policy_loss.py ├── sample_strategy │ ├── __init__.py │ ├── mix_sample_strategy.py │ ├── sample_strategy.py │ └── utils.py └── utils.py ├── buffer ├── __init__.py ├── buffer.py ├── buffer_reader.py ├── buffer_writer.py ├── operators │ ├── __init__.py │ ├── data_juicer_operator.py │ ├── experience_operator.py │ ├── filters │ │ ├── __init__.py │ │ └── reward_filter.py │ └── mappers │ │ ├── __init__.py │ │ ├── pass_rate_calculator.py │ │ └── reward_shaping_mapper.py ├── pipelines │ ├── __init__.py │ ├── experience_pipeline.py │ └── task_pipeline.py ├── reader │ ├── __init__.py │ ├── file_reader.py │ ├── queue_reader.py │ ├── reader.py │ └── sql_reader.py ├── schema │ ├── __init__.py │ ├── formatter.py │ └── sql_schema.py ├── selector │ ├── __init__.py │ ├── difficulty_estimator.py │ └── selector.py ├── storage │ ├── __init__.py │ ├── file.py │ ├── queue.py │ └── sql.py ├── task_scheduler.py ├── utils.py ├── viewer.py └── writer │ ├── __init__.py │ ├── file_writer.py │ ├── queue_writer.py │ └── sql_writer.py ├── cli ├── client.py ├── launcher.py └── server.py ├── common ├── __init__.py ├── config.py ├── constants.py ├── experience.py ├── models │ ├── __init__.py │ ├── mm_utils.py │ ├── model.py │ ├── utils.py │ ├── vllm_model.py │ ├── vllm_patch │ │ ├── __init__.py │ │ ├── api_patch.py │ │ └── worker_patch.py │ └── vllm_worker.py ├── rewards │ ├── __init__.py │ ├── accuracy_reward.py │ ├── agents_reward.py │ ├── countdown_reward.py │ ├── dapo_reward.py │ ├── format_reward.py │ ├── human_reward.py │ ├── math_reward.py │ ├── reward_fn.py │ ├── tool_reward.py │ └── utils.py ├── verl_config.py └── workflows │ ├── __init__.py │ ├── agentscope │ ├── __init__.py │ └── react │ │ ├── __init__.py │ │ ├── react_agent.py │ │ ├── react_workflow.py │ │ └── templates.py │ ├── agentscope_workflow.py │ ├── customized_math_workflows.py │ ├── customized_toolcall_workflows.py │ ├── envs │ ├── agentscope │ │ ├── agentscopev0_react_workflow.py │ │ ├── agentscopev1_react_workflow.py │ │ └── agentscopev1_search_workflow.py │ ├── alfworld │ │ ├── RAFT_alfworld_workflow.py │ │ ├── RAFT_prompt │ │ │ ├── alfworld_system.j2 │ │ │ └── second_attempt_guidance.j2 │ │ ├── RAFT_reflect_alfworld_workflow.py │ │ ├── RAFT_utils.py │ │ └── alfworld_workflow.py │ ├── email_searcher │ │ ├── prepare_data.py │ │ ├── react_agent.py │ │ ├── utils.py │ │ └── workflow.py │ ├── frozen_lake │ │ ├── utils.py │ │ └── workflow.py │ ├── sciworld │ │ └── sciworld_workflow.py │ └── webshop │ │ └── webshop_workflow.py │ ├── eval_workflow.py │ ├── math_rm_workflow.py │ ├── math_ruler_workflow.py │ ├── math_trainable_ruler_workflow.py │ ├── rubric_judge_workflow.py │ ├── simple_mm_workflow.py │ ├── step_wise_workflow.py │ └── workflow.py ├── explorer ├── __init__.py ├── api │ ├── __init__.py │ ├── api.py │ └── service.py ├── explorer.py ├── explorer_client.py ├── scheduler.py └── workflow_runner.py ├── manager ├── __init__.py ├── config_manager.py ├── config_registry │ ├── __init__.py │ ├── algorithm_config_manager.py │ ├── buffer_config_manager.py │ ├── config_registry.py │ ├── explorer_config_manager.py │ ├── model_config_manager.py │ └── trainer_config_manager.py ├── state_manager.py └── synchronizer.py ├── plugins └── __init__.py ├── service ├── __init__.py └── data_juicer │ ├── __init__.py │ ├── client.py │ └── server │ ├── __init__.py │ ├── server.py │ ├── session.py │ └── utils.py ├── trainer ├── __init__.py ├── trainer.py ├── verl │ ├── __init__.py │ ├── dp_actor.py │ ├── fsdp_checkpoint_manager.py │ ├── fsdp_workers.py │ ├── megatron_actor.py │ ├── megatron_checkpoint_manager.py │ ├── megatron_workers.py │ └── utils.py └── verl_trainer.py └── utils ├── __init__.py ├── annotations.py ├── distributed.py ├── dlc_utils.py ├── eval_utils.py ├── log.py ├── lora_utils.py ├── math_eval_utils.py ├── monitor.py ├── plugin_loader.py ├── registry.py └── timer.py /.flake8: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/.flake8 -------------------------------------------------------------------------------- /.github/ISSUE_TEMPLATE/bug_report.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/.github/ISSUE_TEMPLATE/bug_report.md -------------------------------------------------------------------------------- /.github/ISSUE_TEMPLATE/custom.md: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /.github/ISSUE_TEMPLATE/feature_request.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/.github/ISSUE_TEMPLATE/feature_request.md -------------------------------------------------------------------------------- /.github/PULL_REQUEST_TEMPLATE.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/.github/PULL_REQUEST_TEMPLATE.md -------------------------------------------------------------------------------- /.github/workflows/docker.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/.github/workflows/docker.yaml -------------------------------------------------------------------------------- /.github/workflows/docker/docker-compose.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/.github/workflows/docker/docker-compose.yaml -------------------------------------------------------------------------------- /.github/workflows/pre-commit.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/.github/workflows/pre-commit.yaml -------------------------------------------------------------------------------- /.github/workflows/release.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/.github/workflows/release.yaml -------------------------------------------------------------------------------- /.github/workflows/sphinx-doc.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/.github/workflows/sphinx-doc.yaml -------------------------------------------------------------------------------- /.github/workflows/unittest.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/.github/workflows/unittest.yaml -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/.gitignore -------------------------------------------------------------------------------- /.pre-commit-config.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/.pre-commit-config.yaml -------------------------------------------------------------------------------- /CONTRIBUTING.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/CONTRIBUTING.md -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/LICENSE -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/README.md -------------------------------------------------------------------------------- /README_zh.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/README_zh.md -------------------------------------------------------------------------------- /benchmark/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/benchmark/README.md -------------------------------------------------------------------------------- /benchmark/bench.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/benchmark/bench.py -------------------------------------------------------------------------------- /benchmark/config/alfworld-template.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/benchmark/config/alfworld-template.yaml -------------------------------------------------------------------------------- /benchmark/config/countdown-template.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/benchmark/config/countdown-template.yaml -------------------------------------------------------------------------------- /benchmark/config/frozenlake-template.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/benchmark/config/frozenlake-template.yaml -------------------------------------------------------------------------------- /benchmark/config/gsm8k-template.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/benchmark/config/gsm8k-template.yaml -------------------------------------------------------------------------------- /benchmark/config/guru_math-template.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/benchmark/config/guru_math-template.yaml -------------------------------------------------------------------------------- /benchmark/plugins/guru_math/naive_dapo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/benchmark/plugins/guru_math/naive_dapo.py -------------------------------------------------------------------------------- /benchmark/plugins/guru_math/reward.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/benchmark/plugins/guru_math/reward.py -------------------------------------------------------------------------------- /benchmark/reports/alfworld.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/benchmark/reports/alfworld.md -------------------------------------------------------------------------------- /benchmark/reports/frozenlake.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/benchmark/reports/frozenlake.md -------------------------------------------------------------------------------- /benchmark/scripts/gen_countdown_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/benchmark/scripts/gen_countdown_data.py -------------------------------------------------------------------------------- /benchmark/scripts/gen_guru_math_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/benchmark/scripts/gen_guru_math_data.py -------------------------------------------------------------------------------- /benchmark/scripts/get_alfworld_full_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/benchmark/scripts/get_alfworld_full_data.py -------------------------------------------------------------------------------- /docs/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/README.md -------------------------------------------------------------------------------- /docs/sphinx_doc/Makefile: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/Makefile -------------------------------------------------------------------------------- /docs/sphinx_doc/_static/logo.svg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/_static/logo.svg -------------------------------------------------------------------------------- /docs/sphinx_doc/_templates/article_header_customized.html: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/_templates/article_header_customized.html -------------------------------------------------------------------------------- /docs/sphinx_doc/_templates/language-switch.html: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/_templates/language-switch.html -------------------------------------------------------------------------------- /docs/sphinx_doc/_templates/package.rst_t: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/_templates/package.rst_t -------------------------------------------------------------------------------- /docs/sphinx_doc/_templates/versions.html: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/_templates/versions.html -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/DYN-NCCL.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/DYN-NCCL.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/DYN-STATEDICT.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/DYN-STATEDICT.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/FIXED-NCCL.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/FIXED-NCCL.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/FIXED-STATEDICT.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/FIXED-STATEDICT.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/RAFT_alfworld_reward_curve.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/RAFT_alfworld_reward_curve.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/agentscope_dapo_qwen3-4B_reward.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/agentscope_dapo_qwen3-4B_reward.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/agentscope_dapo_reward.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/agentscope_dapo_reward.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/agentscope_dapo_turns.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/agentscope_dapo_turns.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/agentscope_gsm8k_reward.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/agentscope_gsm8k_reward.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/agentscope_gsm8k_turns.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/agentscope_gsm8k_turns.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/agentscope_websearch_reward.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/agentscope_websearch_reward.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/agentscope_websearch_turns.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/agentscope_websearch_turns.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/alfworld_reward_curve.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/alfworld_reward_curve.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/alfworldv2_reward.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/alfworldv2_reward.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/async-curve.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/async-curve.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/bench_alfworld_step.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/bench_alfworld_step.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/bench_alfworld_time.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/bench_alfworld_time.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/bench_frozenlake_step.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/bench_frozenlake_step.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/config-manager.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/config-manager.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/countdown-bench.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/countdown-bench.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/data-in-labelling-page.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/data-in-labelling-page.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/data-in-projects.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/data-in-projects.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/data-labelling.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/data-labelling.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/data-projects.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/data-projects.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/email_eval_accuracy.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/email_eval_accuracy.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/email_reward_mean.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/email_reward_mean.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/email_rollout_accuracy.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/email_rollout_accuracy.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/example_experience_replay/exp_replay_X_explore_step.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/example_experience_replay/exp_replay_X_explore_step.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/example_experience_replay/exp_replay_X_time.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/example_experience_replay/exp_replay_X_time.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/example_experience_replay/exp_replay_model_version.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/example_experience_replay/exp_replay_model_version.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/geometry3k_qwen25_vl_3b_reward.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/geometry3k_qwen25_vl_3b_reward.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/grpo_rubric_reward.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/grpo_rubric_reward.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/gsm8k-bench.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/gsm8k-bench.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/gsm8k_ruler_eval_accuracy.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/gsm8k_ruler_eval_accuracy.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/gsm8k_ruler_gold_reward.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/gsm8k_ruler_gold_reward.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/gsm8k_ruler_judge_success.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/gsm8k_ruler_judge_success.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/gsm8k_ruler_reward.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/gsm8k_ruler_reward.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/gsm8k_trainable_ruler_eval_accuracy.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/gsm8k_trainable_ruler_eval_accuracy.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/gsm8k_trainable_ruler_gold_reward.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/gsm8k_trainable_ruler_gold_reward.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/gsm8k_trainable_ruler_judge_success.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/gsm8k_trainable_ruler_judge_success.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/gsm8k_trainable_ruler_reward.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/gsm8k_trainable_ruler_reward.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/gsm8k_trainable_ruler_reward_for_judger.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/gsm8k_trainable_ruler_reward_for_judger.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/guru-bench.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/guru-bench.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/mix_vlm_reward.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/mix_vlm_reward.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/opmd-curve.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/opmd-curve.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/scripts-multi-plot.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/scripts-multi-plot.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/toolace_3b_response.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/toolace_3b_response.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/toolace_3b_rewards.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/toolace_3b_rewards.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/toolace_length_curve.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/toolace_length_curve.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/toolace_reward_curve.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/toolace_reward_curve.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/trinity-architecture.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/trinity-architecture.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/trinity-design.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/trinity-design.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/trinity-mix.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/trinity-mix.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/trinity-mode.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/trinity-mode.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/trinity-multi-step-compare.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/trinity-multi-step-compare.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/trinity-title.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/trinity-title.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/trinity_data_process.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/trinity_data_process.png -------------------------------------------------------------------------------- /docs/sphinx_doc/assets/webshop_reward_curve.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/assets/webshop_reward_curve.png -------------------------------------------------------------------------------- /docs/sphinx_doc/build_doc.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/build_doc.sh -------------------------------------------------------------------------------- /docs/sphinx_doc/make.bat: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/make.bat -------------------------------------------------------------------------------- /docs/sphinx_doc/source/api_reference.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/api_reference.rst -------------------------------------------------------------------------------- /docs/sphinx_doc/source/conf.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/conf.py -------------------------------------------------------------------------------- /docs/sphinx_doc/source/index.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/index.rst -------------------------------------------------------------------------------- /docs/sphinx_doc/source/main.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/main.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source/tutorial/develop_algorithm.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/tutorial/develop_algorithm.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source/tutorial/develop_operator.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/tutorial/develop_operator.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source/tutorial/develop_overview.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/tutorial/develop_overview.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source/tutorial/develop_selector.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/tutorial/develop_selector.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source/tutorial/develop_workflow.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/tutorial/develop_workflow.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source/tutorial/example_async_mode.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/tutorial/example_async_mode.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source/tutorial/example_data_functionalities.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/tutorial/example_data_functionalities.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source/tutorial/example_dpo.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/tutorial/example_dpo.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source/tutorial/example_megatron.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/tutorial/example_megatron.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source/tutorial/example_mix_algo.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/tutorial/example_mix_algo.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source/tutorial/example_multi_turn.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/tutorial/example_multi_turn.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source/tutorial/example_react.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/tutorial/example_react.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source/tutorial/example_reasoning_advanced.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/tutorial/example_reasoning_advanced.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source/tutorial/example_reasoning_basic.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/tutorial/example_reasoning_basic.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source/tutorial/example_search_email.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/tutorial/example_search_email.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source/tutorial/example_step_wise.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/tutorial/example_step_wise.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source/tutorial/faq.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/tutorial/faq.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source/tutorial/synchronizer.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/tutorial/synchronizer.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source/tutorial/trinity_configs.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/tutorial/trinity_configs.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source/tutorial/trinity_installation.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source/tutorial/trinity_installation.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/api_reference.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/api_reference.rst -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/conf.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/conf.py -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/index.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/index.rst -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/main.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/main.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/tutorial/develop_algorithm.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/tutorial/develop_algorithm.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/tutorial/develop_operator.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/tutorial/develop_operator.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/tutorial/develop_overview.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/tutorial/develop_overview.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/tutorial/develop_selector.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/tutorial/develop_selector.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/tutorial/develop_workflow.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/tutorial/develop_workflow.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/tutorial/example_async_mode.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/tutorial/example_async_mode.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/tutorial/example_data_functionalities.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/tutorial/example_data_functionalities.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/tutorial/example_dpo.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/tutorial/example_dpo.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/tutorial/example_megatron.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/tutorial/example_megatron.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/tutorial/example_mix_algo.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/tutorial/example_mix_algo.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/tutorial/example_multi_turn.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/tutorial/example_multi_turn.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/tutorial/example_react.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/tutorial/example_react.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/tutorial/example_reasoning_advanced.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/tutorial/example_reasoning_advanced.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/tutorial/example_reasoning_basic.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/tutorial/example_reasoning_basic.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/tutorial/example_search_email.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/tutorial/example_search_email.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/tutorial/example_step_wise.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/tutorial/example_step_wise.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/tutorial/faq.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/tutorial/faq.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/tutorial/synchronizer.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/tutorial/synchronizer.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/tutorial/trinity_configs.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/tutorial/trinity_configs.md -------------------------------------------------------------------------------- /docs/sphinx_doc/source_zh/tutorial/trinity_installation.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/docs/sphinx_doc/source_zh/tutorial/trinity_installation.md -------------------------------------------------------------------------------- /environments/data.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/environments/data.yaml -------------------------------------------------------------------------------- /environments/env_mapping.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/environments/env_mapping.json -------------------------------------------------------------------------------- /environments/training.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/environments/training.yaml -------------------------------------------------------------------------------- /examples/RAFT_alfworld/RAFT_alfworld_7B.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/RAFT_alfworld/RAFT_alfworld_7B.yaml -------------------------------------------------------------------------------- /examples/RAFT_alfworld/RAFT_reflect_alfworld_7B.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/RAFT_alfworld/RAFT_reflect_alfworld_7B.yaml -------------------------------------------------------------------------------- /examples/RAFT_alfworld/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/RAFT_alfworld/README.md -------------------------------------------------------------------------------- /examples/agentscope_react/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/agentscope_react/README.md -------------------------------------------------------------------------------- /examples/agentscope_react/gsm8k.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/agentscope_react/gsm8k.yaml -------------------------------------------------------------------------------- /examples/agentscope_tool_react/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/agentscope_tool_react/README.md -------------------------------------------------------------------------------- /examples/agentscope_tool_react/agentscopev0_tool_react_dapo.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/agentscope_tool_react/agentscopev0_tool_react_dapo.yaml -------------------------------------------------------------------------------- /examples/agentscope_tool_react/agentscopev0_tool_react_gsm8k.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/agentscope_tool_react/agentscopev0_tool_react_gsm8k.yaml -------------------------------------------------------------------------------- /examples/agentscope_tool_react/agentscopev1_tool_react_dapo.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/agentscope_tool_react/agentscopev1_tool_react_dapo.yaml -------------------------------------------------------------------------------- /examples/agentscope_websearch/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/agentscope_websearch/README.md -------------------------------------------------------------------------------- /examples/agentscope_websearch/agentscopev1_websearch_agent.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/agentscope_websearch/agentscopev1_websearch_agent.yaml -------------------------------------------------------------------------------- /examples/agentscope_websearch/get_webwalkerQA_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/agentscope_websearch/get_webwalkerQA_data.py -------------------------------------------------------------------------------- /examples/asymre_gsm8k/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/asymre_gsm8k/README.md -------------------------------------------------------------------------------- /examples/asymre_gsm8k/gsm8k.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/asymre_gsm8k/gsm8k.yaml -------------------------------------------------------------------------------- /examples/asymre_math/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/asymre_math/README.md -------------------------------------------------------------------------------- /examples/asymre_math/math.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/asymre_math/math.yaml -------------------------------------------------------------------------------- /examples/async_gsm8k/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/async_gsm8k/README.md -------------------------------------------------------------------------------- /examples/async_gsm8k/explorer.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/async_gsm8k/explorer.yaml -------------------------------------------------------------------------------- /examples/async_gsm8k/run.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/async_gsm8k/run.sh -------------------------------------------------------------------------------- /examples/async_gsm8k/trainer.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/async_gsm8k/trainer.yaml -------------------------------------------------------------------------------- /examples/bots/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/bots/README.md -------------------------------------------------------------------------------- /examples/bots/README_zh.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/bots/README_zh.md -------------------------------------------------------------------------------- /examples/bots/bots.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/bots/bots.yaml -------------------------------------------------------------------------------- /examples/bots/random.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/bots/random.yaml -------------------------------------------------------------------------------- /examples/bots/workflow/bots_math_boxed_reward.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/bots/workflow/bots_math_boxed_reward.py -------------------------------------------------------------------------------- /examples/bots/workflow/bots_math_boxed_workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/bots/workflow/bots_math_boxed_workflow.py -------------------------------------------------------------------------------- /examples/bots/workflow/bots_reward.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/bots/workflow/bots_reward.py -------------------------------------------------------------------------------- /examples/cispo_gsm8k/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/cispo_gsm8k/README.md -------------------------------------------------------------------------------- /examples/cispo_gsm8k/gsm8k.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/cispo_gsm8k/gsm8k.yaml -------------------------------------------------------------------------------- /examples/dapo_math/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/dapo_math/README.md -------------------------------------------------------------------------------- /examples/dapo_math/dapo.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/dapo_math/dapo.yaml -------------------------------------------------------------------------------- /examples/dpo_human_in_the_loop/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/dpo_human_in_the_loop/README.md -------------------------------------------------------------------------------- /examples/dpo_human_in_the_loop/demo-data.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/dpo_human_in_the_loop/demo-data.jsonl -------------------------------------------------------------------------------- /examples/dpo_human_in_the_loop/dpo.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/dpo_human_in_the_loop/dpo.yaml -------------------------------------------------------------------------------- /examples/dpo_humanlike/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/dpo_humanlike/README.md -------------------------------------------------------------------------------- /examples/dpo_humanlike/dpo.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/dpo_humanlike/dpo.yaml -------------------------------------------------------------------------------- /examples/grpo_alfworld/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_alfworld/README.md -------------------------------------------------------------------------------- /examples/grpo_alfworld/alfworld.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_alfworld/alfworld.yaml -------------------------------------------------------------------------------- /examples/grpo_alfworld/get_alfworld_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_alfworld/get_alfworld_data.py -------------------------------------------------------------------------------- /examples/grpo_alfworld_general_multi_step/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_alfworld_general_multi_step/README.md -------------------------------------------------------------------------------- /examples/grpo_alfworld_general_multi_step/alfworld.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_alfworld_general_multi_step/alfworld.yaml -------------------------------------------------------------------------------- /examples/grpo_email_search/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_email_search/README.md -------------------------------------------------------------------------------- /examples/grpo_email_search/email_search.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_email_search/email_search.yaml -------------------------------------------------------------------------------- /examples/grpo_frozen_lake/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_frozen_lake/README.md -------------------------------------------------------------------------------- /examples/grpo_frozen_lake/frozen_lake.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_frozen_lake/frozen_lake.yaml -------------------------------------------------------------------------------- /examples/grpo_frozen_lake/frozen_lake_reward.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_frozen_lake/frozen_lake_reward.png -------------------------------------------------------------------------------- /examples/grpo_frozen_lake/frozen_lake_test_score.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_frozen_lake/frozen_lake_test_score.png -------------------------------------------------------------------------------- /examples/grpo_frozen_lake/get_frozen_lake_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_frozen_lake/get_frozen_lake_data.py -------------------------------------------------------------------------------- /examples/grpo_gsm8k/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_gsm8k/README.md -------------------------------------------------------------------------------- /examples/grpo_gsm8k/gsm8k.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_gsm8k/gsm8k.yaml -------------------------------------------------------------------------------- /examples/grpo_gsm8k_experience_pipeline/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_gsm8k_experience_pipeline/README.md -------------------------------------------------------------------------------- /examples/grpo_gsm8k_experience_pipeline/dj_scoring_exp.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_gsm8k_experience_pipeline/dj_scoring_exp.yaml -------------------------------------------------------------------------------- /examples/grpo_gsm8k_experience_pipeline/gsm8k.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_gsm8k_experience_pipeline/gsm8k.yaml -------------------------------------------------------------------------------- /examples/grpo_gsm8k_ruler/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_gsm8k_ruler/README.md -------------------------------------------------------------------------------- /examples/grpo_gsm8k_ruler/gsm8k_ruler.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_gsm8k_ruler/gsm8k_ruler.yaml -------------------------------------------------------------------------------- /examples/grpo_gsm8k_task_pipeline/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_gsm8k_task_pipeline/README.md -------------------------------------------------------------------------------- /examples/grpo_gsm8k_task_pipeline/gsm8k.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_gsm8k_task_pipeline/gsm8k.yaml -------------------------------------------------------------------------------- /examples/grpo_gsm8k_trainable_ruler/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_gsm8k_trainable_ruler/README.md -------------------------------------------------------------------------------- /examples/grpo_gsm8k_trainable_ruler/gsm8k_ruler.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_gsm8k_trainable_ruler/gsm8k_ruler.yaml -------------------------------------------------------------------------------- /examples/grpo_lora_gsm8k/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_lora_gsm8k/README.md -------------------------------------------------------------------------------- /examples/grpo_lora_gsm8k/gsm8k.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_lora_gsm8k/gsm8k.yaml -------------------------------------------------------------------------------- /examples/grpo_math/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_math/README.md -------------------------------------------------------------------------------- /examples/grpo_math/math.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_math/math.yaml -------------------------------------------------------------------------------- /examples/grpo_rubric_as_reward/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_rubric_as_reward/README.md -------------------------------------------------------------------------------- /examples/grpo_rubric_as_reward/rubric.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_rubric_as_reward/rubric.yaml -------------------------------------------------------------------------------- /examples/grpo_sciworld/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_sciworld/README.md -------------------------------------------------------------------------------- /examples/grpo_sciworld/get_sciworld_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_sciworld/get_sciworld_data.py -------------------------------------------------------------------------------- /examples/grpo_sciworld/sciworld.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_sciworld/sciworld.yaml -------------------------------------------------------------------------------- /examples/grpo_toolcall/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_toolcall/README.md -------------------------------------------------------------------------------- /examples/grpo_toolcall/get_toolace_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_toolcall/get_toolace_data.py -------------------------------------------------------------------------------- /examples/grpo_toolcall/toolace.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_toolcall/toolace.yaml -------------------------------------------------------------------------------- /examples/grpo_vlm/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_vlm/README.md -------------------------------------------------------------------------------- /examples/grpo_vlm/vlm.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_vlm/vlm.yaml -------------------------------------------------------------------------------- /examples/grpo_webshop/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_webshop/README.md -------------------------------------------------------------------------------- /examples/grpo_webshop/get_webshop_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_webshop/get_webshop_data.py -------------------------------------------------------------------------------- /examples/grpo_webshop/webshop.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/grpo_webshop/webshop.yaml -------------------------------------------------------------------------------- /examples/learn_to_ask/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/learn_to_ask/README.md -------------------------------------------------------------------------------- /examples/learn_to_ask/data_prepare/1_info_extract_pipeline.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/learn_to_ask/data_prepare/1_info_extract_pipeline.py -------------------------------------------------------------------------------- /examples/learn_to_ask/data_prepare/2_build_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/learn_to_ask/data_prepare/2_build_dataset.py -------------------------------------------------------------------------------- /examples/learn_to_ask/data_prepare/3_rollout_then_evaluate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/learn_to_ask/data_prepare/3_rollout_then_evaluate.py -------------------------------------------------------------------------------- /examples/learn_to_ask/data_prepare/llm_info_extraction.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/learn_to_ask/data_prepare/llm_info_extraction.py -------------------------------------------------------------------------------- /examples/learn_to_ask/data_prepare/message_splitter.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/learn_to_ask/data_prepare/message_splitter.py -------------------------------------------------------------------------------- /examples/learn_to_ask/train.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/learn_to_ask/train.yaml -------------------------------------------------------------------------------- /examples/learn_to_ask/workflow/prompt_learn2ask.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/learn_to_ask/workflow/prompt_learn2ask.py -------------------------------------------------------------------------------- /examples/learn_to_ask/workflow/workflow_learn2ask.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/learn_to_ask/workflow/workflow_learn2ask.py -------------------------------------------------------------------------------- /examples/mix_chord/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/mix_chord/README.md -------------------------------------------------------------------------------- /examples/mix_chord/eval_bfcl/bfcl_reason_support.patch: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/mix_chord/eval_bfcl/bfcl_reason_support.patch -------------------------------------------------------------------------------- /examples/mix_chord/eval_bfcl/view_single_model_result.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/mix_chord/eval_bfcl/view_single_model_result.py -------------------------------------------------------------------------------- /examples/mix_chord/get_openr1_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/mix_chord/get_openr1_data.py -------------------------------------------------------------------------------- /examples/mix_chord/mix_chord.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/mix_chord/mix_chord.yaml -------------------------------------------------------------------------------- /examples/mix_chord/mix_chord_toolace.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/mix_chord/mix_chord_toolace.yaml -------------------------------------------------------------------------------- /examples/mix_math/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/mix_math/README.md -------------------------------------------------------------------------------- /examples/mix_math/mix_math.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/mix_math/mix_math.yaml -------------------------------------------------------------------------------- /examples/mix_vlm/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/mix_vlm/README.md -------------------------------------------------------------------------------- /examples/mix_vlm/mix_vlm.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/mix_vlm/mix_vlm.yaml -------------------------------------------------------------------------------- /examples/opmd_gsm8k/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/opmd_gsm8k/README.md -------------------------------------------------------------------------------- /examples/opmd_gsm8k/opmd_gsm8k.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/opmd_gsm8k/opmd_gsm8k.yaml -------------------------------------------------------------------------------- /examples/ppo_countdown/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/ppo_countdown/README.md -------------------------------------------------------------------------------- /examples/ppo_countdown/countdown.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/ppo_countdown/countdown.yaml -------------------------------------------------------------------------------- /examples/ppo_countdown_exp_replay/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/ppo_countdown_exp_replay/README.md -------------------------------------------------------------------------------- /examples/ppo_countdown_exp_replay/countdown.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/ppo_countdown_exp_replay/countdown.yaml -------------------------------------------------------------------------------- /examples/ppo_countdown_megatron/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/ppo_countdown_megatron/README.md -------------------------------------------------------------------------------- /examples/ppo_countdown_megatron/countdown.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/ppo_countdown_megatron/countdown.yaml -------------------------------------------------------------------------------- /examples/rec_gsm8k/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/rec_gsm8k/README.md -------------------------------------------------------------------------------- /examples/rec_gsm8k/gsm8k.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/rec_gsm8k/gsm8k.yaml -------------------------------------------------------------------------------- /examples/sft_mot/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/sft_mot/README.md -------------------------------------------------------------------------------- /examples/sft_mot/sft.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/sft_mot/sft.yaml -------------------------------------------------------------------------------- /examples/sppo_gsm8k/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/sppo_gsm8k/README.md -------------------------------------------------------------------------------- /examples/sppo_gsm8k/gsm8k.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/sppo_gsm8k/gsm8k.yaml -------------------------------------------------------------------------------- /examples/topr_gsm8k/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/topr_gsm8k/README.md -------------------------------------------------------------------------------- /examples/topr_gsm8k/gsm8k.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/examples/topr_gsm8k/gsm8k.yaml -------------------------------------------------------------------------------- /pyproject.toml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/pyproject.toml -------------------------------------------------------------------------------- /scripts/docker/Dockerfile: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/scripts/docker/Dockerfile -------------------------------------------------------------------------------- /scripts/docker/Dockerfile.megatron: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/scripts/docker/Dockerfile.megatron -------------------------------------------------------------------------------- /scripts/docker/Dockerfile.uv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/scripts/docker/Dockerfile.uv -------------------------------------------------------------------------------- /scripts/install.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/scripts/install.py -------------------------------------------------------------------------------- /scripts/multi_exps_plot/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/scripts/multi_exps_plot/README.md -------------------------------------------------------------------------------- /scripts/multi_exps_plot/multi_exps_plot.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/scripts/multi_exps_plot/multi_exps_plot.py -------------------------------------------------------------------------------- /scripts/multi_exps_plot/plot_configs.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/scripts/multi_exps_plot/plot_configs.yaml -------------------------------------------------------------------------------- /scripts/start_servers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/scripts/start_servers.py -------------------------------------------------------------------------------- /setup.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/setup.py -------------------------------------------------------------------------------- /tests/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/algorithm/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/algorithm/advantage_fn_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/algorithm/advantage_fn_test.py -------------------------------------------------------------------------------- /tests/algorithm/kl_fn_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/algorithm/kl_fn_test.py -------------------------------------------------------------------------------- /tests/algorithm/policy_loss_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/algorithm/policy_loss_test.py -------------------------------------------------------------------------------- /tests/buffer/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/buffer/experience_pipeline_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/buffer/experience_pipeline_test.py -------------------------------------------------------------------------------- /tests/buffer/experience_storage_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/buffer/experience_storage_test.py -------------------------------------------------------------------------------- /tests/buffer/file_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/buffer/file_test.py -------------------------------------------------------------------------------- /tests/buffer/formatter_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/buffer/formatter_test.py -------------------------------------------------------------------------------- /tests/buffer/queue_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/buffer/queue_test.py -------------------------------------------------------------------------------- /tests/buffer/reader_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/buffer/reader_test.py -------------------------------------------------------------------------------- /tests/buffer/reward_shaping_mapper_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/buffer/reward_shaping_mapper_test.py -------------------------------------------------------------------------------- /tests/buffer/sql_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/buffer/sql_test.py -------------------------------------------------------------------------------- /tests/buffer/task_scheduler_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/buffer/task_scheduler_test.py -------------------------------------------------------------------------------- /tests/buffer/task_storage_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/buffer/task_storage_test.py -------------------------------------------------------------------------------- /tests/cli/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/cli/launcher_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/cli/launcher_test.py -------------------------------------------------------------------------------- /tests/common/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/common/config_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/common/config_test.py -------------------------------------------------------------------------------- /tests/common/experience_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/common/experience_test.py -------------------------------------------------------------------------------- /tests/common/vllm_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/common/vllm_test.py -------------------------------------------------------------------------------- /tests/explorer/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/explorer/explorer_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/explorer/explorer_test.py -------------------------------------------------------------------------------- /tests/explorer/scheduler_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/explorer/scheduler_test.py -------------------------------------------------------------------------------- /tests/explorer/step_wise_workflow_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/explorer/step_wise_workflow_test.py -------------------------------------------------------------------------------- /tests/explorer/workflow_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/explorer/workflow_test.py -------------------------------------------------------------------------------- /tests/manager/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/manager/data/train.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/manager/data/train.jsonl -------------------------------------------------------------------------------- /tests/manager/synchronizer_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/manager/synchronizer_test.py -------------------------------------------------------------------------------- /tests/service/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/service/data_juicer_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/service/data_juicer_test.py -------------------------------------------------------------------------------- /tests/template/config.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/template/config.yaml -------------------------------------------------------------------------------- /tests/template/custom_chat_template.j2: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/template/custom_chat_template.j2 -------------------------------------------------------------------------------- /tests/template/data/countdown/test.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/template/data/countdown/test.jsonl -------------------------------------------------------------------------------- /tests/template/data/countdown/train.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/template/data/countdown/train.jsonl -------------------------------------------------------------------------------- /tests/template/data/eval_long/test.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/template/data/eval_long/test.jsonl -------------------------------------------------------------------------------- /tests/template/data/eval_short/test.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/template/data/eval_short/test.jsonl -------------------------------------------------------------------------------- /tests/template/data/geometry/train.parquet: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/template/data/geometry/train.parquet -------------------------------------------------------------------------------- /tests/template/data/gsm8k/train.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/template/data/gsm8k/train.jsonl -------------------------------------------------------------------------------- /tests/template/data/human_like/train.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/template/data/human_like/train.jsonl -------------------------------------------------------------------------------- /tests/template/data/sft_for_gsm8k/sft.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/template/data/sft_for_gsm8k/sft.jsonl -------------------------------------------------------------------------------- /tests/template/data/sft_with_tools/sft_with_tools.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/template/data/sft_with_tools/sft_with_tools.jsonl -------------------------------------------------------------------------------- /tests/template/data/task_scheduler/subset_1/train.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/template/data/task_scheduler/subset_1/train.jsonl -------------------------------------------------------------------------------- /tests/template/data/task_scheduler/subset_2/train.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/template/data/task_scheduler/subset_2/train.jsonl -------------------------------------------------------------------------------- /tests/tools.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/tools.py -------------------------------------------------------------------------------- /tests/trainer/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/trainer/trainer_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/trainer/trainer_test.py -------------------------------------------------------------------------------- /tests/utils/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/utils/eval_utils_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/utils/eval_utils_test.py -------------------------------------------------------------------------------- /tests/utils/log_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/utils/log_test.py -------------------------------------------------------------------------------- /tests/utils/plugin_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/utils/plugin_test.py -------------------------------------------------------------------------------- /tests/utils/plugins/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/utils/plugins/dependencies.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/utils/plugins/dependencies.py -------------------------------------------------------------------------------- /tests/utils/plugins/main.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/utils/plugins/main.py -------------------------------------------------------------------------------- /tests/utils/plugins/my_monitor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/utils/plugins/my_monitor.py -------------------------------------------------------------------------------- /tests/utils/plugins/my_workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/utils/plugins/my_workflow.py -------------------------------------------------------------------------------- /tests/utils/registry_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/tests/utils/registry_test.py -------------------------------------------------------------------------------- /trinity/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/__init__.py -------------------------------------------------------------------------------- /trinity/algorithm/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/__init__.py -------------------------------------------------------------------------------- /trinity/algorithm/advantage_fn/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/advantage_fn/__init__.py -------------------------------------------------------------------------------- /trinity/algorithm/advantage_fn/advantage_fn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/advantage_fn/advantage_fn.py -------------------------------------------------------------------------------- /trinity/algorithm/advantage_fn/asymre_advantage.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/advantage_fn/asymre_advantage.py -------------------------------------------------------------------------------- /trinity/algorithm/advantage_fn/grpo_advantage.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/advantage_fn/grpo_advantage.py -------------------------------------------------------------------------------- /trinity/algorithm/advantage_fn/multi_step_grpo_advantage.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/advantage_fn/multi_step_grpo_advantage.py -------------------------------------------------------------------------------- /trinity/algorithm/advantage_fn/opmd_advantage.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/advantage_fn/opmd_advantage.py -------------------------------------------------------------------------------- /trinity/algorithm/advantage_fn/ppo_advantage.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/advantage_fn/ppo_advantage.py -------------------------------------------------------------------------------- /trinity/algorithm/advantage_fn/rec_advantage.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/advantage_fn/rec_advantage.py -------------------------------------------------------------------------------- /trinity/algorithm/advantage_fn/reinforce_advantage.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/advantage_fn/reinforce_advantage.py -------------------------------------------------------------------------------- /trinity/algorithm/advantage_fn/reinforce_plus_plus_advantage.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/advantage_fn/reinforce_plus_plus_advantage.py -------------------------------------------------------------------------------- /trinity/algorithm/advantage_fn/remax_advantage.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/advantage_fn/remax_advantage.py -------------------------------------------------------------------------------- /trinity/algorithm/advantage_fn/rloo_advantage.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/advantage_fn/rloo_advantage.py -------------------------------------------------------------------------------- /trinity/algorithm/algorithm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/algorithm.py -------------------------------------------------------------------------------- /trinity/algorithm/entropy_loss_fn/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/entropy_loss_fn/__init__.py -------------------------------------------------------------------------------- /trinity/algorithm/entropy_loss_fn/entropy_loss_fn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/entropy_loss_fn/entropy_loss_fn.py -------------------------------------------------------------------------------- /trinity/algorithm/key_mapper.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/key_mapper.py -------------------------------------------------------------------------------- /trinity/algorithm/kl_fn/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/kl_fn/__init__.py -------------------------------------------------------------------------------- /trinity/algorithm/kl_fn/kl_fn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/kl_fn/kl_fn.py -------------------------------------------------------------------------------- /trinity/algorithm/policy_loss_fn/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/policy_loss_fn/__init__.py -------------------------------------------------------------------------------- /trinity/algorithm/policy_loss_fn/chord_policy_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/policy_loss_fn/chord_policy_loss.py -------------------------------------------------------------------------------- /trinity/algorithm/policy_loss_fn/cispo_policy_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/policy_loss_fn/cispo_policy_loss.py -------------------------------------------------------------------------------- /trinity/algorithm/policy_loss_fn/dpo_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/policy_loss_fn/dpo_loss.py -------------------------------------------------------------------------------- /trinity/algorithm/policy_loss_fn/gspo_policy_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/policy_loss_fn/gspo_policy_loss.py -------------------------------------------------------------------------------- /trinity/algorithm/policy_loss_fn/mix_policy_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/policy_loss_fn/mix_policy_loss.py -------------------------------------------------------------------------------- /trinity/algorithm/policy_loss_fn/opmd_policy_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/policy_loss_fn/opmd_policy_loss.py -------------------------------------------------------------------------------- /trinity/algorithm/policy_loss_fn/policy_loss_fn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/policy_loss_fn/policy_loss_fn.py -------------------------------------------------------------------------------- /trinity/algorithm/policy_loss_fn/ppo_policy_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/policy_loss_fn/ppo_policy_loss.py -------------------------------------------------------------------------------- /trinity/algorithm/policy_loss_fn/rec_policy_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/policy_loss_fn/rec_policy_loss.py -------------------------------------------------------------------------------- /trinity/algorithm/policy_loss_fn/sapo_policy_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/policy_loss_fn/sapo_policy_loss.py -------------------------------------------------------------------------------- /trinity/algorithm/policy_loss_fn/sft_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/policy_loss_fn/sft_loss.py -------------------------------------------------------------------------------- /trinity/algorithm/policy_loss_fn/sppo_loss_fn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/policy_loss_fn/sppo_loss_fn.py -------------------------------------------------------------------------------- /trinity/algorithm/policy_loss_fn/topr_policy_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/policy_loss_fn/topr_policy_loss.py -------------------------------------------------------------------------------- /trinity/algorithm/sample_strategy/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/sample_strategy/__init__.py -------------------------------------------------------------------------------- /trinity/algorithm/sample_strategy/mix_sample_strategy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/sample_strategy/mix_sample_strategy.py -------------------------------------------------------------------------------- /trinity/algorithm/sample_strategy/sample_strategy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/sample_strategy/sample_strategy.py -------------------------------------------------------------------------------- /trinity/algorithm/sample_strategy/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/sample_strategy/utils.py -------------------------------------------------------------------------------- /trinity/algorithm/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/algorithm/utils.py -------------------------------------------------------------------------------- /trinity/buffer/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/__init__.py -------------------------------------------------------------------------------- /trinity/buffer/buffer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/buffer.py -------------------------------------------------------------------------------- /trinity/buffer/buffer_reader.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/buffer_reader.py -------------------------------------------------------------------------------- /trinity/buffer/buffer_writer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/buffer_writer.py -------------------------------------------------------------------------------- /trinity/buffer/operators/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/operators/__init__.py -------------------------------------------------------------------------------- /trinity/buffer/operators/data_juicer_operator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/operators/data_juicer_operator.py -------------------------------------------------------------------------------- /trinity/buffer/operators/experience_operator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/operators/experience_operator.py -------------------------------------------------------------------------------- /trinity/buffer/operators/filters/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /trinity/buffer/operators/filters/reward_filter.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/operators/filters/reward_filter.py -------------------------------------------------------------------------------- /trinity/buffer/operators/mappers/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /trinity/buffer/operators/mappers/pass_rate_calculator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/operators/mappers/pass_rate_calculator.py -------------------------------------------------------------------------------- /trinity/buffer/operators/mappers/reward_shaping_mapper.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/operators/mappers/reward_shaping_mapper.py -------------------------------------------------------------------------------- /trinity/buffer/pipelines/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/pipelines/__init__.py -------------------------------------------------------------------------------- /trinity/buffer/pipelines/experience_pipeline.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/pipelines/experience_pipeline.py -------------------------------------------------------------------------------- /trinity/buffer/pipelines/task_pipeline.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/pipelines/task_pipeline.py -------------------------------------------------------------------------------- /trinity/buffer/reader/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/reader/__init__.py -------------------------------------------------------------------------------- /trinity/buffer/reader/file_reader.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/reader/file_reader.py -------------------------------------------------------------------------------- /trinity/buffer/reader/queue_reader.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/reader/queue_reader.py -------------------------------------------------------------------------------- /trinity/buffer/reader/reader.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/reader/reader.py -------------------------------------------------------------------------------- /trinity/buffer/reader/sql_reader.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/reader/sql_reader.py -------------------------------------------------------------------------------- /trinity/buffer/schema/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/schema/__init__.py -------------------------------------------------------------------------------- /trinity/buffer/schema/formatter.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/schema/formatter.py -------------------------------------------------------------------------------- /trinity/buffer/schema/sql_schema.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/schema/sql_schema.py -------------------------------------------------------------------------------- /trinity/buffer/selector/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/selector/__init__.py -------------------------------------------------------------------------------- /trinity/buffer/selector/difficulty_estimator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/selector/difficulty_estimator.py -------------------------------------------------------------------------------- /trinity/buffer/selector/selector.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/selector/selector.py -------------------------------------------------------------------------------- /trinity/buffer/storage/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /trinity/buffer/storage/file.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/storage/file.py -------------------------------------------------------------------------------- /trinity/buffer/storage/queue.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/storage/queue.py -------------------------------------------------------------------------------- /trinity/buffer/storage/sql.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/storage/sql.py -------------------------------------------------------------------------------- /trinity/buffer/task_scheduler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/task_scheduler.py -------------------------------------------------------------------------------- /trinity/buffer/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/utils.py -------------------------------------------------------------------------------- /trinity/buffer/viewer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/viewer.py -------------------------------------------------------------------------------- /trinity/buffer/writer/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /trinity/buffer/writer/file_writer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/writer/file_writer.py -------------------------------------------------------------------------------- /trinity/buffer/writer/queue_writer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/writer/queue_writer.py -------------------------------------------------------------------------------- /trinity/buffer/writer/sql_writer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/buffer/writer/sql_writer.py -------------------------------------------------------------------------------- /trinity/cli/client.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/cli/client.py -------------------------------------------------------------------------------- /trinity/cli/launcher.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/cli/launcher.py -------------------------------------------------------------------------------- /trinity/cli/server.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/cli/server.py -------------------------------------------------------------------------------- /trinity/common/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /trinity/common/config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/config.py -------------------------------------------------------------------------------- /trinity/common/constants.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/constants.py -------------------------------------------------------------------------------- /trinity/common/experience.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/experience.py -------------------------------------------------------------------------------- /trinity/common/models/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/models/__init__.py -------------------------------------------------------------------------------- /trinity/common/models/mm_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/models/mm_utils.py -------------------------------------------------------------------------------- /trinity/common/models/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/models/model.py -------------------------------------------------------------------------------- /trinity/common/models/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/models/utils.py -------------------------------------------------------------------------------- /trinity/common/models/vllm_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/models/vllm_model.py -------------------------------------------------------------------------------- /trinity/common/models/vllm_patch/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/models/vllm_patch/__init__.py -------------------------------------------------------------------------------- /trinity/common/models/vllm_patch/api_patch.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/models/vllm_patch/api_patch.py -------------------------------------------------------------------------------- /trinity/common/models/vllm_patch/worker_patch.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/models/vllm_patch/worker_patch.py -------------------------------------------------------------------------------- /trinity/common/models/vllm_worker.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/models/vllm_worker.py -------------------------------------------------------------------------------- /trinity/common/rewards/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/rewards/__init__.py -------------------------------------------------------------------------------- /trinity/common/rewards/accuracy_reward.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/rewards/accuracy_reward.py -------------------------------------------------------------------------------- /trinity/common/rewards/agents_reward.py: -------------------------------------------------------------------------------- 1 | # to be implemented 2 | -------------------------------------------------------------------------------- /trinity/common/rewards/countdown_reward.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/rewards/countdown_reward.py -------------------------------------------------------------------------------- /trinity/common/rewards/dapo_reward.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/rewards/dapo_reward.py -------------------------------------------------------------------------------- /trinity/common/rewards/format_reward.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/rewards/format_reward.py -------------------------------------------------------------------------------- /trinity/common/rewards/human_reward.py: -------------------------------------------------------------------------------- 1 | # to be implemented 2 | -------------------------------------------------------------------------------- /trinity/common/rewards/math_reward.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/rewards/math_reward.py -------------------------------------------------------------------------------- /trinity/common/rewards/reward_fn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/rewards/reward_fn.py -------------------------------------------------------------------------------- /trinity/common/rewards/tool_reward.py: -------------------------------------------------------------------------------- 1 | # to be implemented 2 | -------------------------------------------------------------------------------- /trinity/common/rewards/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/rewards/utils.py -------------------------------------------------------------------------------- /trinity/common/verl_config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/verl_config.py -------------------------------------------------------------------------------- /trinity/common/workflows/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/__init__.py -------------------------------------------------------------------------------- /trinity/common/workflows/agentscope/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/agentscope/__init__.py -------------------------------------------------------------------------------- /trinity/common/workflows/agentscope/react/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /trinity/common/workflows/agentscope/react/react_agent.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/agentscope/react/react_agent.py -------------------------------------------------------------------------------- /trinity/common/workflows/agentscope/react/react_workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/agentscope/react/react_workflow.py -------------------------------------------------------------------------------- /trinity/common/workflows/agentscope/react/templates.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/agentscope/react/templates.py -------------------------------------------------------------------------------- /trinity/common/workflows/agentscope_workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/agentscope_workflow.py -------------------------------------------------------------------------------- /trinity/common/workflows/customized_math_workflows.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/customized_math_workflows.py -------------------------------------------------------------------------------- /trinity/common/workflows/customized_toolcall_workflows.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/customized_toolcall_workflows.py -------------------------------------------------------------------------------- /trinity/common/workflows/envs/agentscope/agentscopev0_react_workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/envs/agentscope/agentscopev0_react_workflow.py -------------------------------------------------------------------------------- /trinity/common/workflows/envs/agentscope/agentscopev1_react_workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/envs/agentscope/agentscopev1_react_workflow.py -------------------------------------------------------------------------------- /trinity/common/workflows/envs/agentscope/agentscopev1_search_workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/envs/agentscope/agentscopev1_search_workflow.py -------------------------------------------------------------------------------- /trinity/common/workflows/envs/alfworld/RAFT_alfworld_workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/envs/alfworld/RAFT_alfworld_workflow.py -------------------------------------------------------------------------------- /trinity/common/workflows/envs/alfworld/RAFT_prompt/alfworld_system.j2: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/envs/alfworld/RAFT_prompt/alfworld_system.j2 -------------------------------------------------------------------------------- /trinity/common/workflows/envs/alfworld/RAFT_prompt/second_attempt_guidance.j2: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/envs/alfworld/RAFT_prompt/second_attempt_guidance.j2 -------------------------------------------------------------------------------- /trinity/common/workflows/envs/alfworld/RAFT_reflect_alfworld_workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/envs/alfworld/RAFT_reflect_alfworld_workflow.py -------------------------------------------------------------------------------- /trinity/common/workflows/envs/alfworld/RAFT_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/envs/alfworld/RAFT_utils.py -------------------------------------------------------------------------------- /trinity/common/workflows/envs/alfworld/alfworld_workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/envs/alfworld/alfworld_workflow.py -------------------------------------------------------------------------------- /trinity/common/workflows/envs/email_searcher/prepare_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/envs/email_searcher/prepare_data.py -------------------------------------------------------------------------------- /trinity/common/workflows/envs/email_searcher/react_agent.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/envs/email_searcher/react_agent.py -------------------------------------------------------------------------------- /trinity/common/workflows/envs/email_searcher/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/envs/email_searcher/utils.py -------------------------------------------------------------------------------- /trinity/common/workflows/envs/email_searcher/workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/envs/email_searcher/workflow.py -------------------------------------------------------------------------------- /trinity/common/workflows/envs/frozen_lake/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/envs/frozen_lake/utils.py -------------------------------------------------------------------------------- /trinity/common/workflows/envs/frozen_lake/workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/envs/frozen_lake/workflow.py -------------------------------------------------------------------------------- /trinity/common/workflows/envs/sciworld/sciworld_workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/envs/sciworld/sciworld_workflow.py -------------------------------------------------------------------------------- /trinity/common/workflows/envs/webshop/webshop_workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/envs/webshop/webshop_workflow.py -------------------------------------------------------------------------------- /trinity/common/workflows/eval_workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/eval_workflow.py -------------------------------------------------------------------------------- /trinity/common/workflows/math_rm_workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/math_rm_workflow.py -------------------------------------------------------------------------------- /trinity/common/workflows/math_ruler_workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/math_ruler_workflow.py -------------------------------------------------------------------------------- /trinity/common/workflows/math_trainable_ruler_workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/math_trainable_ruler_workflow.py -------------------------------------------------------------------------------- /trinity/common/workflows/rubric_judge_workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/rubric_judge_workflow.py -------------------------------------------------------------------------------- /trinity/common/workflows/simple_mm_workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/simple_mm_workflow.py -------------------------------------------------------------------------------- /trinity/common/workflows/step_wise_workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/step_wise_workflow.py -------------------------------------------------------------------------------- /trinity/common/workflows/workflow.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/common/workflows/workflow.py -------------------------------------------------------------------------------- /trinity/explorer/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/explorer/__init__.py -------------------------------------------------------------------------------- /trinity/explorer/api/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /trinity/explorer/api/api.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/explorer/api/api.py -------------------------------------------------------------------------------- /trinity/explorer/api/service.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/explorer/api/service.py -------------------------------------------------------------------------------- /trinity/explorer/explorer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/explorer/explorer.py -------------------------------------------------------------------------------- /trinity/explorer/explorer_client.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/explorer/explorer_client.py -------------------------------------------------------------------------------- /trinity/explorer/scheduler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/explorer/scheduler.py -------------------------------------------------------------------------------- /trinity/explorer/workflow_runner.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/explorer/workflow_runner.py -------------------------------------------------------------------------------- /trinity/manager/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/manager/__init__.py -------------------------------------------------------------------------------- /trinity/manager/config_manager.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/manager/config_manager.py -------------------------------------------------------------------------------- /trinity/manager/config_registry/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/manager/config_registry/__init__.py -------------------------------------------------------------------------------- /trinity/manager/config_registry/algorithm_config_manager.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/manager/config_registry/algorithm_config_manager.py -------------------------------------------------------------------------------- /trinity/manager/config_registry/buffer_config_manager.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/manager/config_registry/buffer_config_manager.py -------------------------------------------------------------------------------- /trinity/manager/config_registry/config_registry.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/manager/config_registry/config_registry.py -------------------------------------------------------------------------------- /trinity/manager/config_registry/explorer_config_manager.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/manager/config_registry/explorer_config_manager.py -------------------------------------------------------------------------------- /trinity/manager/config_registry/model_config_manager.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/manager/config_registry/model_config_manager.py -------------------------------------------------------------------------------- /trinity/manager/config_registry/trainer_config_manager.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/manager/config_registry/trainer_config_manager.py -------------------------------------------------------------------------------- /trinity/manager/state_manager.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/manager/state_manager.py -------------------------------------------------------------------------------- /trinity/manager/synchronizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/manager/synchronizer.py -------------------------------------------------------------------------------- /trinity/plugins/__init__.py: -------------------------------------------------------------------------------- 1 | """Add your custom modules to this directory.""" 2 | -------------------------------------------------------------------------------- /trinity/service/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/service/__init__.py -------------------------------------------------------------------------------- /trinity/service/data_juicer/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /trinity/service/data_juicer/client.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/service/data_juicer/client.py -------------------------------------------------------------------------------- /trinity/service/data_juicer/server/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /trinity/service/data_juicer/server/server.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/service/data_juicer/server/server.py -------------------------------------------------------------------------------- /trinity/service/data_juicer/server/session.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/service/data_juicer/server/session.py -------------------------------------------------------------------------------- /trinity/service/data_juicer/server/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/service/data_juicer/server/utils.py -------------------------------------------------------------------------------- /trinity/trainer/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/trainer/__init__.py -------------------------------------------------------------------------------- /trinity/trainer/trainer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/trainer/trainer.py -------------------------------------------------------------------------------- /trinity/trainer/verl/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /trinity/trainer/verl/dp_actor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/trainer/verl/dp_actor.py -------------------------------------------------------------------------------- /trinity/trainer/verl/fsdp_checkpoint_manager.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/trainer/verl/fsdp_checkpoint_manager.py -------------------------------------------------------------------------------- /trinity/trainer/verl/fsdp_workers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/trainer/verl/fsdp_workers.py -------------------------------------------------------------------------------- /trinity/trainer/verl/megatron_actor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/trainer/verl/megatron_actor.py -------------------------------------------------------------------------------- /trinity/trainer/verl/megatron_checkpoint_manager.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/trainer/verl/megatron_checkpoint_manager.py -------------------------------------------------------------------------------- /trinity/trainer/verl/megatron_workers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/trainer/verl/megatron_workers.py -------------------------------------------------------------------------------- /trinity/trainer/verl/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/trainer/verl/utils.py -------------------------------------------------------------------------------- /trinity/trainer/verl_trainer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/trainer/verl_trainer.py -------------------------------------------------------------------------------- /trinity/utils/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /trinity/utils/annotations.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/utils/annotations.py -------------------------------------------------------------------------------- /trinity/utils/distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/utils/distributed.py -------------------------------------------------------------------------------- /trinity/utils/dlc_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/utils/dlc_utils.py -------------------------------------------------------------------------------- /trinity/utils/eval_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/utils/eval_utils.py -------------------------------------------------------------------------------- /trinity/utils/log.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/utils/log.py -------------------------------------------------------------------------------- /trinity/utils/lora_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/utils/lora_utils.py -------------------------------------------------------------------------------- /trinity/utils/math_eval_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/utils/math_eval_utils.py -------------------------------------------------------------------------------- /trinity/utils/monitor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/utils/monitor.py -------------------------------------------------------------------------------- /trinity/utils/plugin_loader.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/utils/plugin_loader.py -------------------------------------------------------------------------------- /trinity/utils/registry.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/utils/registry.py -------------------------------------------------------------------------------- /trinity/utils/timer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/modelscope/Trinity-RFT/HEAD/trinity/utils/timer.py --------------------------------------------------------------------------------