├── .github └── workflows │ ├── lint.yml │ └── main.yml ├── .gitignore ├── .gitmodules ├── .pre-commit-config.yaml ├── Dockerfile ├── Dockerfile_cu124 ├── LICENSE ├── README.md ├── README_zh.md ├── assets ├── wan_i2v │ ├── calib │ │ ├── astronaut.jpg │ │ └── samples.json │ └── eval │ │ ├── astronaut.jpg │ │ └── samples.json └── wan_t2v │ ├── calib │ └── samples.json │ └── eval │ └── samples.json ├── ci_check ├── awq_w4a16_fakequant_eval.yml ├── change_files.py ├── cpu.txt ├── gptq_w_only.yml ├── model_urls.txt ├── run_awq.sh └── run_gptq.sh ├── configs ├── opencompass │ ├── eval_base.py │ └── eval_chat.py ├── quantization │ ├── backend │ │ ├── autoawq │ │ │ ├── awq_w4a16.yml │ │ │ ├── gptq_w4a16.yml │ │ │ ├── rtn_w4a16.yml │ │ │ └── w4a16_combin │ │ │ │ ├── step_1_awq.yml │ │ │ │ └── step_2_omniq.yml │ │ ├── mlcllm │ │ │ ├── awq_w4a16.yml │ │ │ ├── gptq_w4a16.yml │ │ │ ├── rtn_w4a16.yml │ │ │ └── w4a16_combin │ │ │ │ ├── step_1_awq.yml │ │ │ │ └── step_2_omniq.yml │ │ ├── sglang │ │ │ ├── awq_w4a16.yml │ │ │ ├── awq_w8a8.yml │ │ │ ├── fp8 │ │ │ │ ├── awq_fp8.yml │ │ │ │ ├── awq_fp8_static.yml │ │ │ │ ├── gptq_fp8.yml │ │ │ │ ├── rtn_fp8.yml │ │ │ │ └── smoothquant_fp8.yml │ │ │ ├── gptq_w4a16.yml │ │ │ ├── rtn_w4a16.yml │ │ │ ├── rtn_w8a16.yml │ │ │ ├── rtn_w8a8.yml │ │ │ ├── smoothquant_w8a8.yml │ │ │ ├── w4a16_combin │ │ │ │ ├── step_1_awq.yml │ │ │ │ └── step_2_omniq.yml │ │ │ └── w8a8_combin │ │ │ │ ├── step_1_quarot.yml │ │ │ │ └── step_2_gptq.yml │ │ ├── trtllm │ │ │ ├── awq_w4a16.yml │ │ │ └── smoothquant_w8a8.yml │ │ └── vllm │ │ │ ├── awq_w4a16.yml │ │ │ ├── awq_w8a8.yml │ │ │ ├── fp8 │ │ │ ├── awq_fp8.yml │ │ │ ├── awq_fp8_static.yml │ │ │ ├── gptq_fp8.yml │ │ │ ├── rtn_fp8.yml │ │ │ └── smoothquant_fp8.yml │ │ │ ├── gptq_w4a16.yml │ │ │ ├── rtn_w4a16.yml │ │ │ ├── rtn_w4a16_dsv3.yml │ │ │ ├── rtn_w8a16.yml │ │ │ ├── rtn_w8a8.yml │ │ │ ├── smoothquant_w8a8.yml │ │ │ ├── tesseraq_w4a16.yml │ │ │ ├── w4a16_combin │ │ │ ├── step_1_awq.yml │ │ │ └── step_2_omniq.yml │ │ │ └── w8a8_combin │ │ │ ├── step_1_quarot.yml │ │ │ └── step_2_gptq.yml │ ├── combination │ │ ├── awq_comb_omni │ │ │ ├── w2a16g64 │ │ │ │ ├── step_1_awq.yml │ │ │ │ └── step_2_omniq.yml │ │ │ ├── w3a16g128 │ │ │ │ ├── step_1_awq.yml │ │ │ │ └── step_2_omniq.yml │ │ │ ├── w4a16g128 │ │ │ │ ├── step_1_awq.yml │ │ │ │ └── step_2_omniq.yml │ │ │ ├── w6a6 │ │ │ │ ├── step_1_awq.yml │ │ │ │ └── step_2_omniq.yml │ │ │ └── w8a8 │ │ │ │ ├── step_1_awq.yml │ │ │ │ └── step_2_omniq.yml │ │ ├── quant_comb_token_pruning │ │ │ └── rtn_w_a_vlm.yml │ │ └── quarot_comb_gptq │ │ │ ├── w4a4 │ │ │ ├── step_1_quarot.yml │ │ │ └── step_2_gptq.yml │ │ │ └── w8a8 │ │ │ ├── step_1_quarot.yml │ │ │ └── step_2_gptq.yml │ ├── deepseekv3 │ │ ├── awq_w_only_dsv3.yml │ │ ├── awq_w_only_dsv3_bf16.yml │ │ ├── osplus_w_a_dsv3.yml │ │ ├── quarot_w_a_dsv3.yml │ │ ├── rtn_w_a_dsv3.yml │ │ ├── rtn_w_only_dsv3.yml │ │ └── smoothquant_w_a_dsv3.yml │ ├── methods │ │ ├── AdaDim │ │ │ └── adadim_w_a.yml │ │ ├── Awq │ │ │ ├── awq_w_a.yml │ │ │ ├── awq_w_a_chat_data.yml │ │ │ ├── awq_w_a_mix_bits.yml │ │ │ ├── awq_w_only.yml │ │ │ ├── awq_w_only_custom_alm_data_padding.yml │ │ │ ├── awq_w_only_custom_avlm_data_padding.yml │ │ │ ├── awq_w_only_custom_data.yml │ │ │ ├── awq_w_only_custom_data_debug.yml │ │ │ ├── awq_w_only_custom_data_padding.yml │ │ │ ├── awq_w_only_custom_eval.yml │ │ │ ├── awq_w_only_custom_vlm_data_padding.yml │ │ │ ├── awq_w_only_mix_bits_1.yml │ │ │ ├── awq_w_only_mix_bits_2.yml │ │ │ ├── awq_w_only_opencompass.yml │ │ │ ├── awq_w_only_vit.yml │ │ │ └── awq_w_only_vlm.yml │ │ ├── DGQ │ │ │ └── dgq_w_a.yml │ │ ├── FP_Quant │ │ │ ├── awq_we2m1a16_g128.yml │ │ │ ├── gptq_we2m1a16_g128.yml │ │ │ ├── rtn_we2m1a16_g128.yml │ │ │ ├── rtn_we2m1ae2m1.yml │ │ │ ├── rtn_we4m3ae4m3.yml │ │ │ └── rtn_we5m2ae5m2.yml │ │ ├── GPTQ │ │ │ ├── gptq_owq_w_only.yml │ │ │ ├── gptq_w_only.yml │ │ │ └── gptq_w_only_vlm.yml │ │ ├── HQQ │ │ │ └── hqq_w_only.yml │ │ ├── KVQuant │ │ │ ├── rtn_w_a_kivi_quant_kv.yml │ │ │ ├── rtn_w_a_naive_quant_kv.yml │ │ │ └── rtn_w_a_pertensor_static_naive_quant_kv.yml │ │ ├── LlmInt8 │ │ │ └── llmint8_w_only.yml │ │ ├── MixPrecision │ │ │ ├── awq_w_a.yml │ │ │ ├── awq_w_a_static.yml │ │ │ ├── rtn_w_a.yml │ │ │ └── rtn_w_a_static.yml │ │ ├── NormTweaking │ │ │ ├── ntweak_w_a.yml │ │ │ └── ntweak_w_only.yml │ │ ├── OmniQuant │ │ │ ├── omniq_w_a.yml │ │ │ └── omniq_w_only.yml │ │ ├── OsPlus │ │ │ └── osplus_w_a.yml │ │ ├── QUIK │ │ │ └── quik_w_a.yml │ │ ├── QuaRot │ │ │ └── quarot_w_a.yml │ │ ├── RTN │ │ │ ├── rtn_w_a.yml │ │ │ ├── rtn_w_a_block.yml │ │ │ ├── rtn_w_a_kv.yml │ │ │ ├── rtn_w_a_pertensor_static.yml │ │ │ ├── rtn_w_a_vlm.yml │ │ │ ├── rtn_w_a_wint4afp8.yml │ │ │ ├── rtn_w_a_wint4aint8.yml │ │ │ └── rtn_w_only.yml │ │ ├── SmoothQuant │ │ │ ├── smoothquant_w_a.yml │ │ │ └── smoothquant_w_a_vlm.yml │ │ ├── SpQR │ │ │ └── spqr_w_only.yml │ │ └── Tesseraq │ │ │ └── tesseraq_w_only.yml │ └── video_gen │ │ ├── wan_i2v │ │ ├── awq_w_a.yaml │ │ ├── rtn_w_a.yaml │ │ ├── rtn_w_a_lora.yaml │ │ ├── smoothquant_w_a.yaml │ │ ├── smoothquant_w_a_fp8.yaml │ │ └── smoothquant_w_a_int8_lora.yaml │ │ └── wan_t2v │ │ ├── awq_w_a.yaml │ │ ├── rtn_w_a.yaml │ │ └── smoothquant_w_a.yaml └── sparsification │ └── methods │ ├── DART │ └── dart.yml │ ├── DivPrune │ └── divprune.yml │ ├── DyCoke │ └── dycoke.yml │ ├── FastV │ └── fastv.yml │ ├── FastVID │ └── fastvid.yml │ ├── FasterVLM │ └── fastervlm.yml │ ├── Holitom │ └── holitom.yml │ ├── Kvsparse │ ├── shadowkv.yml │ └── sinkkv.yml │ ├── Magnitude │ └── magnitude.yml │ ├── MustDrop │ └── mustdrop.yml │ ├── PruneVid │ └── prunevid.yml │ ├── PyramidDrop │ └── pyramidrop.yml │ ├── ShortGPT │ └── shortgpt.yml │ ├── SparseVLM │ ├── sparsevlm.yml │ └── sparsevlm_multi_turn.yml │ ├── ToMe │ └── tome.yml │ ├── VisPruner │ └── vispruner.yml │ ├── VisionZip │ └── visionzip.yml │ └── Wanda │ └── wanda.yml ├── docs ├── en │ ├── .readthedocs.yaml │ ├── Makefile │ ├── make.bat │ └── source │ │ ├── advanced │ │ ├── VLM_quant&img-txt_dataset.md │ │ ├── Vit_quant&img_dataset.md │ │ ├── custom_dataset.md │ │ ├── mix_bits.md │ │ ├── model_test_v1.md │ │ ├── model_test_v2.md │ │ ├── sparsification.md │ │ └── token_reduction.md │ │ ├── backend │ │ ├── autoawq.md │ │ ├── lightx2v.md │ │ ├── mlcllm.md │ │ ├── sglang.md │ │ └── vllm.md │ │ ├── conf.py │ │ ├── configs.md │ │ ├── images │ │ └── logo │ │ │ └── llmc.svg │ │ ├── index.rst │ │ ├── practice │ │ ├── awq.md │ │ ├── awq_omni.md │ │ └── quarot_gptq.md │ │ └── quickstart.md └── zh_cn │ ├── .readthedocs.yaml │ ├── Makefile │ ├── make.bat │ └── source │ ├── advanced │ ├── VLM_quant&img-txt_dataset.md │ ├── Vit_quant&img_dataset.md │ ├── custom_dataset.md │ ├── mix_bits.md │ ├── model_test_v1.md │ ├── model_test_v2.md │ ├── sparsification.md │ └── token_reduction.md │ ├── backend │ ├── autoawq.md │ ├── lightx2v.md │ ├── mlcllm.md │ ├── sglang.md │ └── vllm.md │ ├── conf.py │ ├── configs.md │ ├── images │ └── logo │ │ └── llmc.svg │ ├── index.rst │ ├── practice │ ├── awq.md │ ├── awq_omni.md │ └── quarot_gptq.md │ └── quickstart.md ├── examples └── backend │ ├── autoawq │ └── infer_with_autoawq.py │ ├── mlcllm │ └── infer_with_mlcllm.py │ ├── sglang │ └── infer_with_sglang.py │ └── vllm │ └── infer_with_vllm.py ├── imgs ├── K.png ├── best_practice.png ├── llmc+.png └── llmc.png ├── llmc ├── __main__.py ├── compression │ ├── __init__.py │ ├── blockwise_optimization.py │ ├── quantization │ │ ├── __init__.py │ │ ├── adadim.py │ │ ├── attn_utils.py │ │ ├── auto_clip.py │ │ ├── awq.py │ │ ├── base_blockwise_quantization.py │ │ ├── dgq.py │ │ ├── gptq.py │ │ ├── hadamard_utils.py │ │ ├── hqq.py │ │ ├── kernel.py │ │ ├── kvquant.py │ │ ├── llmint8.py │ │ ├── module_utils.py │ │ ├── ntweak.py │ │ ├── omniq.py │ │ ├── osplus.py │ │ ├── quant.py │ │ ├── quarot.py │ │ ├── quik.py │ │ ├── rtn.py │ │ ├── smoothquant.py │ │ ├── spqr.py │ │ ├── tesseraq.py │ │ ├── train_utils.py │ │ └── utils.py │ ├── sparsification │ │ ├── __init__.py │ │ ├── attn_utils.py │ │ ├── base_blockwise_sparsification.py │ │ ├── dense.py │ │ ├── kvsparse.py │ │ ├── magnitude.py │ │ ├── shortgpt.py │ │ └── wanda.py │ └── token_reduction │ │ ├── __init__.py │ │ ├── base_blockwise_token_reduction.py │ │ ├── dart.py │ │ ├── divprune.py │ │ ├── dycoke.py │ │ ├── fastervlm.py │ │ ├── fastv.py │ │ ├── fastvid.py │ │ ├── holitom.py │ │ ├── mustdrop.py │ │ ├── prunevid.py │ │ ├── pyramiddrop.py │ │ ├── random.py │ │ ├── sparsevlm.py │ │ ├── token_reduction_module.py │ │ ├── tome.py │ │ ├── utils.py │ │ ├── visionzip.py │ │ ├── vispruner.py │ │ └── visualizer.py ├── data │ ├── __init__.py │ └── dataset │ │ ├── __init__.py │ │ ├── base_dataset.py │ │ └── specified_preproc.py ├── eval │ ├── __init__.py │ ├── eval_acc.py │ ├── eval_base.py │ ├── eval_code.py │ ├── eval_custom_generate.py │ ├── eval_custom_generate_just_infer.py │ ├── eval_ppl.py │ ├── eval_token_consist.py │ ├── eval_video_generate.py │ ├── eval_vqa.py │ └── utils.py ├── models │ ├── __init__.py │ ├── base_model.py │ ├── bloom.py │ ├── chatglm.py │ ├── deepseekv2.py │ ├── deepseekv3.py │ ├── falcon.py │ ├── gemma2.py │ ├── glm4v.py │ ├── internlm2.py │ ├── internomni.py │ ├── internvl2.py │ ├── internvl3_5.py │ ├── llama.py │ ├── llava.py │ ├── llava_hf.py │ ├── llava_onevision.py │ ├── minicpm.py │ ├── minicpmv.py │ ├── mistral.py │ ├── mixtral.py │ ├── mllama.py │ ├── opt.py │ ├── phi.py │ ├── phi3.py │ ├── qwen.py │ ├── qwen2.py │ ├── qwen2_5vl.py │ ├── qwen2audio.py │ ├── qwen2moe.py │ ├── qwen2vl.py │ ├── qwen3.py │ ├── qwen3moe.py │ ├── smollm.py │ ├── stablelm.py │ ├── starcoder.py │ ├── videollava.py │ ├── vila.py │ ├── vit.py │ ├── wan_i2v.py │ └── wan_t2v.py └── utils │ ├── __init__.py │ ├── export_autoawq.py │ ├── export_lightx2v.py │ ├── export_trtllm.py │ ├── export_vllm.py │ ├── registry_factory.py │ ├── utils.py │ └── visualizer.py ├── requirements.txt ├── requirements ├── docs.txt └── runtime.txt ├── scripts ├── run_llmc.sh └── run_lm_eval.sh └── tools ├── attention_vis.py ├── download_calib_dataset.py ├── download_eval_dataset.py ├── llm_eval.py ├── print_model.py └── quant_analysis.py /.github/workflows/lint.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/.github/workflows/lint.yml -------------------------------------------------------------------------------- /.github/workflows/main.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/.github/workflows/main.yml -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/.gitignore -------------------------------------------------------------------------------- /.gitmodules: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/.gitmodules -------------------------------------------------------------------------------- /.pre-commit-config.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/.pre-commit-config.yaml -------------------------------------------------------------------------------- /Dockerfile: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/Dockerfile -------------------------------------------------------------------------------- /Dockerfile_cu124: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/Dockerfile_cu124 -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/LICENSE -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/README.md -------------------------------------------------------------------------------- /README_zh.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/README_zh.md -------------------------------------------------------------------------------- /assets/wan_i2v/calib/astronaut.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/assets/wan_i2v/calib/astronaut.jpg -------------------------------------------------------------------------------- /assets/wan_i2v/calib/samples.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/assets/wan_i2v/calib/samples.json -------------------------------------------------------------------------------- /assets/wan_i2v/eval/astronaut.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/assets/wan_i2v/eval/astronaut.jpg -------------------------------------------------------------------------------- /assets/wan_i2v/eval/samples.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/assets/wan_i2v/eval/samples.json -------------------------------------------------------------------------------- /assets/wan_t2v/calib/samples.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/assets/wan_t2v/calib/samples.json -------------------------------------------------------------------------------- /assets/wan_t2v/eval/samples.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/assets/wan_t2v/eval/samples.json -------------------------------------------------------------------------------- /ci_check/awq_w4a16_fakequant_eval.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/ci_check/awq_w4a16_fakequant_eval.yml -------------------------------------------------------------------------------- /ci_check/change_files.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/ci_check/change_files.py -------------------------------------------------------------------------------- /ci_check/cpu.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/ci_check/cpu.txt -------------------------------------------------------------------------------- /ci_check/gptq_w_only.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/ci_check/gptq_w_only.yml -------------------------------------------------------------------------------- /ci_check/model_urls.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/ci_check/model_urls.txt -------------------------------------------------------------------------------- /ci_check/run_awq.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/ci_check/run_awq.sh -------------------------------------------------------------------------------- /ci_check/run_gptq.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/ci_check/run_gptq.sh -------------------------------------------------------------------------------- /configs/opencompass/eval_base.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/opencompass/eval_base.py -------------------------------------------------------------------------------- /configs/opencompass/eval_chat.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/opencompass/eval_chat.py -------------------------------------------------------------------------------- /configs/quantization/backend/autoawq/awq_w4a16.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/autoawq/awq_w4a16.yml -------------------------------------------------------------------------------- /configs/quantization/backend/autoawq/gptq_w4a16.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/autoawq/gptq_w4a16.yml -------------------------------------------------------------------------------- /configs/quantization/backend/autoawq/rtn_w4a16.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/autoawq/rtn_w4a16.yml -------------------------------------------------------------------------------- /configs/quantization/backend/autoawq/w4a16_combin/step_1_awq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/autoawq/w4a16_combin/step_1_awq.yml -------------------------------------------------------------------------------- /configs/quantization/backend/autoawq/w4a16_combin/step_2_omniq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/autoawq/w4a16_combin/step_2_omniq.yml -------------------------------------------------------------------------------- /configs/quantization/backend/mlcllm/awq_w4a16.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/mlcllm/awq_w4a16.yml -------------------------------------------------------------------------------- /configs/quantization/backend/mlcllm/gptq_w4a16.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/mlcllm/gptq_w4a16.yml -------------------------------------------------------------------------------- /configs/quantization/backend/mlcllm/rtn_w4a16.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/mlcllm/rtn_w4a16.yml -------------------------------------------------------------------------------- /configs/quantization/backend/mlcllm/w4a16_combin/step_1_awq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/mlcllm/w4a16_combin/step_1_awq.yml -------------------------------------------------------------------------------- /configs/quantization/backend/mlcllm/w4a16_combin/step_2_omniq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/mlcllm/w4a16_combin/step_2_omniq.yml -------------------------------------------------------------------------------- /configs/quantization/backend/sglang/awq_w4a16.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/sglang/awq_w4a16.yml -------------------------------------------------------------------------------- /configs/quantization/backend/sglang/awq_w8a8.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/sglang/awq_w8a8.yml -------------------------------------------------------------------------------- /configs/quantization/backend/sglang/fp8/awq_fp8.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/sglang/fp8/awq_fp8.yml -------------------------------------------------------------------------------- /configs/quantization/backend/sglang/fp8/awq_fp8_static.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/sglang/fp8/awq_fp8_static.yml -------------------------------------------------------------------------------- /configs/quantization/backend/sglang/fp8/gptq_fp8.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/sglang/fp8/gptq_fp8.yml -------------------------------------------------------------------------------- /configs/quantization/backend/sglang/fp8/rtn_fp8.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/sglang/fp8/rtn_fp8.yml -------------------------------------------------------------------------------- /configs/quantization/backend/sglang/fp8/smoothquant_fp8.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/sglang/fp8/smoothquant_fp8.yml -------------------------------------------------------------------------------- /configs/quantization/backend/sglang/gptq_w4a16.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/sglang/gptq_w4a16.yml -------------------------------------------------------------------------------- /configs/quantization/backend/sglang/rtn_w4a16.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/sglang/rtn_w4a16.yml -------------------------------------------------------------------------------- /configs/quantization/backend/sglang/rtn_w8a16.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/sglang/rtn_w8a16.yml -------------------------------------------------------------------------------- /configs/quantization/backend/sglang/rtn_w8a8.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/sglang/rtn_w8a8.yml -------------------------------------------------------------------------------- /configs/quantization/backend/sglang/smoothquant_w8a8.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/sglang/smoothquant_w8a8.yml -------------------------------------------------------------------------------- /configs/quantization/backend/sglang/w4a16_combin/step_1_awq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/sglang/w4a16_combin/step_1_awq.yml -------------------------------------------------------------------------------- /configs/quantization/backend/sglang/w4a16_combin/step_2_omniq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/sglang/w4a16_combin/step_2_omniq.yml -------------------------------------------------------------------------------- /configs/quantization/backend/sglang/w8a8_combin/step_1_quarot.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/sglang/w8a8_combin/step_1_quarot.yml -------------------------------------------------------------------------------- /configs/quantization/backend/sglang/w8a8_combin/step_2_gptq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/sglang/w8a8_combin/step_2_gptq.yml -------------------------------------------------------------------------------- /configs/quantization/backend/trtllm/awq_w4a16.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/trtllm/awq_w4a16.yml -------------------------------------------------------------------------------- /configs/quantization/backend/trtllm/smoothquant_w8a8.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/trtllm/smoothquant_w8a8.yml -------------------------------------------------------------------------------- /configs/quantization/backend/vllm/awq_w4a16.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/vllm/awq_w4a16.yml -------------------------------------------------------------------------------- /configs/quantization/backend/vllm/awq_w8a8.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/vllm/awq_w8a8.yml -------------------------------------------------------------------------------- /configs/quantization/backend/vllm/fp8/awq_fp8.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/vllm/fp8/awq_fp8.yml -------------------------------------------------------------------------------- /configs/quantization/backend/vllm/fp8/awq_fp8_static.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/vllm/fp8/awq_fp8_static.yml -------------------------------------------------------------------------------- /configs/quantization/backend/vllm/fp8/gptq_fp8.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/vllm/fp8/gptq_fp8.yml -------------------------------------------------------------------------------- /configs/quantization/backend/vllm/fp8/rtn_fp8.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/vllm/fp8/rtn_fp8.yml -------------------------------------------------------------------------------- /configs/quantization/backend/vllm/fp8/smoothquant_fp8.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/vllm/fp8/smoothquant_fp8.yml -------------------------------------------------------------------------------- /configs/quantization/backend/vllm/gptq_w4a16.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/vllm/gptq_w4a16.yml -------------------------------------------------------------------------------- /configs/quantization/backend/vllm/rtn_w4a16.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/vllm/rtn_w4a16.yml -------------------------------------------------------------------------------- /configs/quantization/backend/vllm/rtn_w4a16_dsv3.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/vllm/rtn_w4a16_dsv3.yml -------------------------------------------------------------------------------- /configs/quantization/backend/vllm/rtn_w8a16.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/vllm/rtn_w8a16.yml -------------------------------------------------------------------------------- /configs/quantization/backend/vllm/rtn_w8a8.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/vllm/rtn_w8a8.yml -------------------------------------------------------------------------------- /configs/quantization/backend/vllm/smoothquant_w8a8.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/vllm/smoothquant_w8a8.yml -------------------------------------------------------------------------------- /configs/quantization/backend/vllm/tesseraq_w4a16.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/vllm/tesseraq_w4a16.yml -------------------------------------------------------------------------------- /configs/quantization/backend/vllm/w4a16_combin/step_1_awq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/vllm/w4a16_combin/step_1_awq.yml -------------------------------------------------------------------------------- /configs/quantization/backend/vllm/w4a16_combin/step_2_omniq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/vllm/w4a16_combin/step_2_omniq.yml -------------------------------------------------------------------------------- /configs/quantization/backend/vllm/w8a8_combin/step_1_quarot.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/vllm/w8a8_combin/step_1_quarot.yml -------------------------------------------------------------------------------- /configs/quantization/backend/vllm/w8a8_combin/step_2_gptq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/backend/vllm/w8a8_combin/step_2_gptq.yml -------------------------------------------------------------------------------- /configs/quantization/combination/awq_comb_omni/w2a16g64/step_1_awq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/combination/awq_comb_omni/w2a16g64/step_1_awq.yml -------------------------------------------------------------------------------- /configs/quantization/combination/awq_comb_omni/w2a16g64/step_2_omniq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/combination/awq_comb_omni/w2a16g64/step_2_omniq.yml -------------------------------------------------------------------------------- /configs/quantization/combination/awq_comb_omni/w3a16g128/step_1_awq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/combination/awq_comb_omni/w3a16g128/step_1_awq.yml -------------------------------------------------------------------------------- /configs/quantization/combination/awq_comb_omni/w3a16g128/step_2_omniq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/combination/awq_comb_omni/w3a16g128/step_2_omniq.yml -------------------------------------------------------------------------------- /configs/quantization/combination/awq_comb_omni/w4a16g128/step_1_awq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/combination/awq_comb_omni/w4a16g128/step_1_awq.yml -------------------------------------------------------------------------------- /configs/quantization/combination/awq_comb_omni/w4a16g128/step_2_omniq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/combination/awq_comb_omni/w4a16g128/step_2_omniq.yml -------------------------------------------------------------------------------- /configs/quantization/combination/awq_comb_omni/w6a6/step_1_awq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/combination/awq_comb_omni/w6a6/step_1_awq.yml -------------------------------------------------------------------------------- /configs/quantization/combination/awq_comb_omni/w6a6/step_2_omniq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/combination/awq_comb_omni/w6a6/step_2_omniq.yml -------------------------------------------------------------------------------- /configs/quantization/combination/awq_comb_omni/w8a8/step_1_awq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/combination/awq_comb_omni/w8a8/step_1_awq.yml -------------------------------------------------------------------------------- /configs/quantization/combination/awq_comb_omni/w8a8/step_2_omniq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/combination/awq_comb_omni/w8a8/step_2_omniq.yml -------------------------------------------------------------------------------- /configs/quantization/combination/quant_comb_token_pruning/rtn_w_a_vlm.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/combination/quant_comb_token_pruning/rtn_w_a_vlm.yml -------------------------------------------------------------------------------- /configs/quantization/combination/quarot_comb_gptq/w4a4/step_1_quarot.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/combination/quarot_comb_gptq/w4a4/step_1_quarot.yml -------------------------------------------------------------------------------- /configs/quantization/combination/quarot_comb_gptq/w4a4/step_2_gptq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/combination/quarot_comb_gptq/w4a4/step_2_gptq.yml -------------------------------------------------------------------------------- /configs/quantization/combination/quarot_comb_gptq/w8a8/step_1_quarot.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/combination/quarot_comb_gptq/w8a8/step_1_quarot.yml -------------------------------------------------------------------------------- /configs/quantization/combination/quarot_comb_gptq/w8a8/step_2_gptq.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/combination/quarot_comb_gptq/w8a8/step_2_gptq.yml -------------------------------------------------------------------------------- /configs/quantization/deepseekv3/awq_w_only_dsv3.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/deepseekv3/awq_w_only_dsv3.yml -------------------------------------------------------------------------------- /configs/quantization/deepseekv3/awq_w_only_dsv3_bf16.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/deepseekv3/awq_w_only_dsv3_bf16.yml -------------------------------------------------------------------------------- /configs/quantization/deepseekv3/osplus_w_a_dsv3.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/deepseekv3/osplus_w_a_dsv3.yml -------------------------------------------------------------------------------- /configs/quantization/deepseekv3/quarot_w_a_dsv3.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/deepseekv3/quarot_w_a_dsv3.yml -------------------------------------------------------------------------------- /configs/quantization/deepseekv3/rtn_w_a_dsv3.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/deepseekv3/rtn_w_a_dsv3.yml -------------------------------------------------------------------------------- /configs/quantization/deepseekv3/rtn_w_only_dsv3.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/deepseekv3/rtn_w_only_dsv3.yml -------------------------------------------------------------------------------- /configs/quantization/deepseekv3/smoothquant_w_a_dsv3.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/deepseekv3/smoothquant_w_a_dsv3.yml -------------------------------------------------------------------------------- /configs/quantization/methods/AdaDim/adadim_w_a.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/AdaDim/adadim_w_a.yml -------------------------------------------------------------------------------- /configs/quantization/methods/Awq/awq_w_a.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/Awq/awq_w_a.yml -------------------------------------------------------------------------------- /configs/quantization/methods/Awq/awq_w_a_chat_data.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/Awq/awq_w_a_chat_data.yml -------------------------------------------------------------------------------- /configs/quantization/methods/Awq/awq_w_a_mix_bits.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/Awq/awq_w_a_mix_bits.yml -------------------------------------------------------------------------------- /configs/quantization/methods/Awq/awq_w_only.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/Awq/awq_w_only.yml -------------------------------------------------------------------------------- /configs/quantization/methods/Awq/awq_w_only_custom_alm_data_padding.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/Awq/awq_w_only_custom_alm_data_padding.yml -------------------------------------------------------------------------------- /configs/quantization/methods/Awq/awq_w_only_custom_avlm_data_padding.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/Awq/awq_w_only_custom_avlm_data_padding.yml -------------------------------------------------------------------------------- /configs/quantization/methods/Awq/awq_w_only_custom_data.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/Awq/awq_w_only_custom_data.yml -------------------------------------------------------------------------------- /configs/quantization/methods/Awq/awq_w_only_custom_data_debug.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/Awq/awq_w_only_custom_data_debug.yml -------------------------------------------------------------------------------- /configs/quantization/methods/Awq/awq_w_only_custom_data_padding.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/Awq/awq_w_only_custom_data_padding.yml -------------------------------------------------------------------------------- /configs/quantization/methods/Awq/awq_w_only_custom_eval.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/Awq/awq_w_only_custom_eval.yml -------------------------------------------------------------------------------- /configs/quantization/methods/Awq/awq_w_only_custom_vlm_data_padding.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/Awq/awq_w_only_custom_vlm_data_padding.yml -------------------------------------------------------------------------------- /configs/quantization/methods/Awq/awq_w_only_mix_bits_1.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/Awq/awq_w_only_mix_bits_1.yml -------------------------------------------------------------------------------- /configs/quantization/methods/Awq/awq_w_only_mix_bits_2.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/Awq/awq_w_only_mix_bits_2.yml -------------------------------------------------------------------------------- /configs/quantization/methods/Awq/awq_w_only_opencompass.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/Awq/awq_w_only_opencompass.yml -------------------------------------------------------------------------------- /configs/quantization/methods/Awq/awq_w_only_vit.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/Awq/awq_w_only_vit.yml -------------------------------------------------------------------------------- /configs/quantization/methods/Awq/awq_w_only_vlm.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/Awq/awq_w_only_vlm.yml -------------------------------------------------------------------------------- /configs/quantization/methods/DGQ/dgq_w_a.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/DGQ/dgq_w_a.yml -------------------------------------------------------------------------------- /configs/quantization/methods/FP_Quant/awq_we2m1a16_g128.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/FP_Quant/awq_we2m1a16_g128.yml -------------------------------------------------------------------------------- /configs/quantization/methods/FP_Quant/gptq_we2m1a16_g128.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/FP_Quant/gptq_we2m1a16_g128.yml -------------------------------------------------------------------------------- /configs/quantization/methods/FP_Quant/rtn_we2m1a16_g128.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/FP_Quant/rtn_we2m1a16_g128.yml -------------------------------------------------------------------------------- /configs/quantization/methods/FP_Quant/rtn_we2m1ae2m1.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/FP_Quant/rtn_we2m1ae2m1.yml -------------------------------------------------------------------------------- /configs/quantization/methods/FP_Quant/rtn_we4m3ae4m3.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/FP_Quant/rtn_we4m3ae4m3.yml -------------------------------------------------------------------------------- /configs/quantization/methods/FP_Quant/rtn_we5m2ae5m2.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/FP_Quant/rtn_we5m2ae5m2.yml -------------------------------------------------------------------------------- /configs/quantization/methods/GPTQ/gptq_owq_w_only.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/GPTQ/gptq_owq_w_only.yml -------------------------------------------------------------------------------- /configs/quantization/methods/GPTQ/gptq_w_only.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/GPTQ/gptq_w_only.yml -------------------------------------------------------------------------------- /configs/quantization/methods/GPTQ/gptq_w_only_vlm.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/GPTQ/gptq_w_only_vlm.yml -------------------------------------------------------------------------------- /configs/quantization/methods/HQQ/hqq_w_only.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/HQQ/hqq_w_only.yml -------------------------------------------------------------------------------- /configs/quantization/methods/KVQuant/rtn_w_a_kivi_quant_kv.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/KVQuant/rtn_w_a_kivi_quant_kv.yml -------------------------------------------------------------------------------- /configs/quantization/methods/KVQuant/rtn_w_a_naive_quant_kv.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/KVQuant/rtn_w_a_naive_quant_kv.yml -------------------------------------------------------------------------------- /configs/quantization/methods/KVQuant/rtn_w_a_pertensor_static_naive_quant_kv.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/KVQuant/rtn_w_a_pertensor_static_naive_quant_kv.yml -------------------------------------------------------------------------------- /configs/quantization/methods/LlmInt8/llmint8_w_only.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/LlmInt8/llmint8_w_only.yml -------------------------------------------------------------------------------- /configs/quantization/methods/MixPrecision/awq_w_a.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/MixPrecision/awq_w_a.yml -------------------------------------------------------------------------------- /configs/quantization/methods/MixPrecision/awq_w_a_static.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/MixPrecision/awq_w_a_static.yml -------------------------------------------------------------------------------- /configs/quantization/methods/MixPrecision/rtn_w_a.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/MixPrecision/rtn_w_a.yml -------------------------------------------------------------------------------- /configs/quantization/methods/MixPrecision/rtn_w_a_static.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/MixPrecision/rtn_w_a_static.yml -------------------------------------------------------------------------------- /configs/quantization/methods/NormTweaking/ntweak_w_a.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/NormTweaking/ntweak_w_a.yml -------------------------------------------------------------------------------- /configs/quantization/methods/NormTweaking/ntweak_w_only.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/NormTweaking/ntweak_w_only.yml -------------------------------------------------------------------------------- /configs/quantization/methods/OmniQuant/omniq_w_a.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/OmniQuant/omniq_w_a.yml -------------------------------------------------------------------------------- /configs/quantization/methods/OmniQuant/omniq_w_only.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/OmniQuant/omniq_w_only.yml -------------------------------------------------------------------------------- /configs/quantization/methods/OsPlus/osplus_w_a.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/OsPlus/osplus_w_a.yml -------------------------------------------------------------------------------- /configs/quantization/methods/QUIK/quik_w_a.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/QUIK/quik_w_a.yml -------------------------------------------------------------------------------- /configs/quantization/methods/QuaRot/quarot_w_a.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/QuaRot/quarot_w_a.yml -------------------------------------------------------------------------------- /configs/quantization/methods/RTN/rtn_w_a.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/RTN/rtn_w_a.yml -------------------------------------------------------------------------------- /configs/quantization/methods/RTN/rtn_w_a_block.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/RTN/rtn_w_a_block.yml -------------------------------------------------------------------------------- /configs/quantization/methods/RTN/rtn_w_a_kv.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/RTN/rtn_w_a_kv.yml -------------------------------------------------------------------------------- /configs/quantization/methods/RTN/rtn_w_a_pertensor_static.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/RTN/rtn_w_a_pertensor_static.yml -------------------------------------------------------------------------------- /configs/quantization/methods/RTN/rtn_w_a_vlm.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/RTN/rtn_w_a_vlm.yml -------------------------------------------------------------------------------- /configs/quantization/methods/RTN/rtn_w_a_wint4afp8.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/RTN/rtn_w_a_wint4afp8.yml -------------------------------------------------------------------------------- /configs/quantization/methods/RTN/rtn_w_a_wint4aint8.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/RTN/rtn_w_a_wint4aint8.yml -------------------------------------------------------------------------------- /configs/quantization/methods/RTN/rtn_w_only.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/RTN/rtn_w_only.yml -------------------------------------------------------------------------------- /configs/quantization/methods/SmoothQuant/smoothquant_w_a.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/SmoothQuant/smoothquant_w_a.yml -------------------------------------------------------------------------------- /configs/quantization/methods/SmoothQuant/smoothquant_w_a_vlm.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/SmoothQuant/smoothquant_w_a_vlm.yml -------------------------------------------------------------------------------- /configs/quantization/methods/SpQR/spqr_w_only.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/SpQR/spqr_w_only.yml -------------------------------------------------------------------------------- /configs/quantization/methods/Tesseraq/tesseraq_w_only.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/methods/Tesseraq/tesseraq_w_only.yml -------------------------------------------------------------------------------- /configs/quantization/video_gen/wan_i2v/awq_w_a.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/video_gen/wan_i2v/awq_w_a.yaml -------------------------------------------------------------------------------- /configs/quantization/video_gen/wan_i2v/rtn_w_a.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/video_gen/wan_i2v/rtn_w_a.yaml -------------------------------------------------------------------------------- /configs/quantization/video_gen/wan_i2v/rtn_w_a_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/video_gen/wan_i2v/rtn_w_a_lora.yaml -------------------------------------------------------------------------------- /configs/quantization/video_gen/wan_i2v/smoothquant_w_a.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/video_gen/wan_i2v/smoothquant_w_a.yaml -------------------------------------------------------------------------------- /configs/quantization/video_gen/wan_i2v/smoothquant_w_a_fp8.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/video_gen/wan_i2v/smoothquant_w_a_fp8.yaml -------------------------------------------------------------------------------- /configs/quantization/video_gen/wan_i2v/smoothquant_w_a_int8_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/video_gen/wan_i2v/smoothquant_w_a_int8_lora.yaml -------------------------------------------------------------------------------- /configs/quantization/video_gen/wan_t2v/awq_w_a.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/video_gen/wan_t2v/awq_w_a.yaml -------------------------------------------------------------------------------- /configs/quantization/video_gen/wan_t2v/rtn_w_a.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/video_gen/wan_t2v/rtn_w_a.yaml -------------------------------------------------------------------------------- /configs/quantization/video_gen/wan_t2v/smoothquant_w_a.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/quantization/video_gen/wan_t2v/smoothquant_w_a.yaml -------------------------------------------------------------------------------- /configs/sparsification/methods/DART/dart.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/sparsification/methods/DART/dart.yml -------------------------------------------------------------------------------- /configs/sparsification/methods/DivPrune/divprune.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/sparsification/methods/DivPrune/divprune.yml -------------------------------------------------------------------------------- /configs/sparsification/methods/DyCoke/dycoke.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/sparsification/methods/DyCoke/dycoke.yml -------------------------------------------------------------------------------- /configs/sparsification/methods/FastV/fastv.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/sparsification/methods/FastV/fastv.yml -------------------------------------------------------------------------------- /configs/sparsification/methods/FastVID/fastvid.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/sparsification/methods/FastVID/fastvid.yml -------------------------------------------------------------------------------- /configs/sparsification/methods/FasterVLM/fastervlm.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/sparsification/methods/FasterVLM/fastervlm.yml -------------------------------------------------------------------------------- /configs/sparsification/methods/Holitom/holitom.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/sparsification/methods/Holitom/holitom.yml -------------------------------------------------------------------------------- /configs/sparsification/methods/Kvsparse/shadowkv.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/sparsification/methods/Kvsparse/shadowkv.yml -------------------------------------------------------------------------------- /configs/sparsification/methods/Kvsparse/sinkkv.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/sparsification/methods/Kvsparse/sinkkv.yml -------------------------------------------------------------------------------- /configs/sparsification/methods/Magnitude/magnitude.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/sparsification/methods/Magnitude/magnitude.yml -------------------------------------------------------------------------------- /configs/sparsification/methods/MustDrop/mustdrop.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/sparsification/methods/MustDrop/mustdrop.yml -------------------------------------------------------------------------------- /configs/sparsification/methods/PruneVid/prunevid.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/sparsification/methods/PruneVid/prunevid.yml -------------------------------------------------------------------------------- /configs/sparsification/methods/PyramidDrop/pyramidrop.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/sparsification/methods/PyramidDrop/pyramidrop.yml -------------------------------------------------------------------------------- /configs/sparsification/methods/ShortGPT/shortgpt.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/sparsification/methods/ShortGPT/shortgpt.yml -------------------------------------------------------------------------------- /configs/sparsification/methods/SparseVLM/sparsevlm.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/sparsification/methods/SparseVLM/sparsevlm.yml -------------------------------------------------------------------------------- /configs/sparsification/methods/SparseVLM/sparsevlm_multi_turn.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/sparsification/methods/SparseVLM/sparsevlm_multi_turn.yml -------------------------------------------------------------------------------- /configs/sparsification/methods/ToMe/tome.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/sparsification/methods/ToMe/tome.yml -------------------------------------------------------------------------------- /configs/sparsification/methods/VisPruner/vispruner.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/sparsification/methods/VisPruner/vispruner.yml -------------------------------------------------------------------------------- /configs/sparsification/methods/VisionZip/visionzip.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/sparsification/methods/VisionZip/visionzip.yml -------------------------------------------------------------------------------- /configs/sparsification/methods/Wanda/wanda.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/configs/sparsification/methods/Wanda/wanda.yml -------------------------------------------------------------------------------- /docs/en/.readthedocs.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/.readthedocs.yaml -------------------------------------------------------------------------------- /docs/en/Makefile: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/Makefile -------------------------------------------------------------------------------- /docs/en/make.bat: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/make.bat -------------------------------------------------------------------------------- /docs/en/source/advanced/VLM_quant&img-txt_dataset.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/source/advanced/VLM_quant&img-txt_dataset.md -------------------------------------------------------------------------------- /docs/en/source/advanced/Vit_quant&img_dataset.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/source/advanced/Vit_quant&img_dataset.md -------------------------------------------------------------------------------- /docs/en/source/advanced/custom_dataset.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/source/advanced/custom_dataset.md -------------------------------------------------------------------------------- /docs/en/source/advanced/mix_bits.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/source/advanced/mix_bits.md -------------------------------------------------------------------------------- /docs/en/source/advanced/model_test_v1.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/source/advanced/model_test_v1.md -------------------------------------------------------------------------------- /docs/en/source/advanced/model_test_v2.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/source/advanced/model_test_v2.md -------------------------------------------------------------------------------- /docs/en/source/advanced/sparsification.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/source/advanced/sparsification.md -------------------------------------------------------------------------------- /docs/en/source/advanced/token_reduction.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/source/advanced/token_reduction.md -------------------------------------------------------------------------------- /docs/en/source/backend/autoawq.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/source/backend/autoawq.md -------------------------------------------------------------------------------- /docs/en/source/backend/lightx2v.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/source/backend/lightx2v.md -------------------------------------------------------------------------------- /docs/en/source/backend/mlcllm.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/source/backend/mlcllm.md -------------------------------------------------------------------------------- /docs/en/source/backend/sglang.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/source/backend/sglang.md -------------------------------------------------------------------------------- /docs/en/source/backend/vllm.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/source/backend/vllm.md -------------------------------------------------------------------------------- /docs/en/source/conf.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/source/conf.py -------------------------------------------------------------------------------- /docs/en/source/configs.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/source/configs.md -------------------------------------------------------------------------------- /docs/en/source/images/logo/llmc.svg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/source/images/logo/llmc.svg -------------------------------------------------------------------------------- /docs/en/source/index.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/source/index.rst -------------------------------------------------------------------------------- /docs/en/source/practice/awq.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/source/practice/awq.md -------------------------------------------------------------------------------- /docs/en/source/practice/awq_omni.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/source/practice/awq_omni.md -------------------------------------------------------------------------------- /docs/en/source/practice/quarot_gptq.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/source/practice/quarot_gptq.md -------------------------------------------------------------------------------- /docs/en/source/quickstart.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/en/source/quickstart.md -------------------------------------------------------------------------------- /docs/zh_cn/.readthedocs.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/.readthedocs.yaml -------------------------------------------------------------------------------- /docs/zh_cn/Makefile: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/Makefile -------------------------------------------------------------------------------- /docs/zh_cn/make.bat: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/make.bat -------------------------------------------------------------------------------- /docs/zh_cn/source/advanced/VLM_quant&img-txt_dataset.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/source/advanced/VLM_quant&img-txt_dataset.md -------------------------------------------------------------------------------- /docs/zh_cn/source/advanced/Vit_quant&img_dataset.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/source/advanced/Vit_quant&img_dataset.md -------------------------------------------------------------------------------- /docs/zh_cn/source/advanced/custom_dataset.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/source/advanced/custom_dataset.md -------------------------------------------------------------------------------- /docs/zh_cn/source/advanced/mix_bits.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/source/advanced/mix_bits.md -------------------------------------------------------------------------------- /docs/zh_cn/source/advanced/model_test_v1.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/source/advanced/model_test_v1.md -------------------------------------------------------------------------------- /docs/zh_cn/source/advanced/model_test_v2.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/source/advanced/model_test_v2.md -------------------------------------------------------------------------------- /docs/zh_cn/source/advanced/sparsification.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/source/advanced/sparsification.md -------------------------------------------------------------------------------- /docs/zh_cn/source/advanced/token_reduction.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/source/advanced/token_reduction.md -------------------------------------------------------------------------------- /docs/zh_cn/source/backend/autoawq.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/source/backend/autoawq.md -------------------------------------------------------------------------------- /docs/zh_cn/source/backend/lightx2v.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/source/backend/lightx2v.md -------------------------------------------------------------------------------- /docs/zh_cn/source/backend/mlcllm.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/source/backend/mlcllm.md -------------------------------------------------------------------------------- /docs/zh_cn/source/backend/sglang.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/source/backend/sglang.md -------------------------------------------------------------------------------- /docs/zh_cn/source/backend/vllm.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/source/backend/vllm.md -------------------------------------------------------------------------------- /docs/zh_cn/source/conf.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/source/conf.py -------------------------------------------------------------------------------- /docs/zh_cn/source/configs.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/source/configs.md -------------------------------------------------------------------------------- /docs/zh_cn/source/images/logo/llmc.svg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/source/images/logo/llmc.svg -------------------------------------------------------------------------------- /docs/zh_cn/source/index.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/source/index.rst -------------------------------------------------------------------------------- /docs/zh_cn/source/practice/awq.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/source/practice/awq.md -------------------------------------------------------------------------------- /docs/zh_cn/source/practice/awq_omni.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/source/practice/awq_omni.md -------------------------------------------------------------------------------- /docs/zh_cn/source/practice/quarot_gptq.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/source/practice/quarot_gptq.md -------------------------------------------------------------------------------- /docs/zh_cn/source/quickstart.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/docs/zh_cn/source/quickstart.md -------------------------------------------------------------------------------- /examples/backend/autoawq/infer_with_autoawq.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/examples/backend/autoawq/infer_with_autoawq.py -------------------------------------------------------------------------------- /examples/backend/mlcllm/infer_with_mlcllm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/examples/backend/mlcllm/infer_with_mlcllm.py -------------------------------------------------------------------------------- /examples/backend/sglang/infer_with_sglang.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/examples/backend/sglang/infer_with_sglang.py -------------------------------------------------------------------------------- /examples/backend/vllm/infer_with_vllm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/examples/backend/vllm/infer_with_vllm.py -------------------------------------------------------------------------------- /imgs/K.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/imgs/K.png -------------------------------------------------------------------------------- /imgs/best_practice.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/imgs/best_practice.png -------------------------------------------------------------------------------- /imgs/llmc+.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/imgs/llmc+.png -------------------------------------------------------------------------------- /imgs/llmc.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/imgs/llmc.png -------------------------------------------------------------------------------- /llmc/__main__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/__main__.py -------------------------------------------------------------------------------- /llmc/compression/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/__init__.py -------------------------------------------------------------------------------- /llmc/compression/blockwise_optimization.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/blockwise_optimization.py -------------------------------------------------------------------------------- /llmc/compression/quantization/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/__init__.py -------------------------------------------------------------------------------- /llmc/compression/quantization/adadim.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/adadim.py -------------------------------------------------------------------------------- /llmc/compression/quantization/attn_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/attn_utils.py -------------------------------------------------------------------------------- /llmc/compression/quantization/auto_clip.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/auto_clip.py -------------------------------------------------------------------------------- /llmc/compression/quantization/awq.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/awq.py -------------------------------------------------------------------------------- /llmc/compression/quantization/base_blockwise_quantization.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/base_blockwise_quantization.py -------------------------------------------------------------------------------- /llmc/compression/quantization/dgq.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/dgq.py -------------------------------------------------------------------------------- /llmc/compression/quantization/gptq.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/gptq.py -------------------------------------------------------------------------------- /llmc/compression/quantization/hadamard_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/hadamard_utils.py -------------------------------------------------------------------------------- /llmc/compression/quantization/hqq.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/hqq.py -------------------------------------------------------------------------------- /llmc/compression/quantization/kernel.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/kernel.py -------------------------------------------------------------------------------- /llmc/compression/quantization/kvquant.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/kvquant.py -------------------------------------------------------------------------------- /llmc/compression/quantization/llmint8.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/llmint8.py -------------------------------------------------------------------------------- /llmc/compression/quantization/module_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/module_utils.py -------------------------------------------------------------------------------- /llmc/compression/quantization/ntweak.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/ntweak.py -------------------------------------------------------------------------------- /llmc/compression/quantization/omniq.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/omniq.py -------------------------------------------------------------------------------- /llmc/compression/quantization/osplus.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/osplus.py -------------------------------------------------------------------------------- /llmc/compression/quantization/quant.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/quant.py -------------------------------------------------------------------------------- /llmc/compression/quantization/quarot.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/quarot.py -------------------------------------------------------------------------------- /llmc/compression/quantization/quik.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/quik.py -------------------------------------------------------------------------------- /llmc/compression/quantization/rtn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/rtn.py -------------------------------------------------------------------------------- /llmc/compression/quantization/smoothquant.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/smoothquant.py -------------------------------------------------------------------------------- /llmc/compression/quantization/spqr.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/spqr.py -------------------------------------------------------------------------------- /llmc/compression/quantization/tesseraq.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/tesseraq.py -------------------------------------------------------------------------------- /llmc/compression/quantization/train_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/train_utils.py -------------------------------------------------------------------------------- /llmc/compression/quantization/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/quantization/utils.py -------------------------------------------------------------------------------- /llmc/compression/sparsification/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/sparsification/__init__.py -------------------------------------------------------------------------------- /llmc/compression/sparsification/attn_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/sparsification/attn_utils.py -------------------------------------------------------------------------------- /llmc/compression/sparsification/base_blockwise_sparsification.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/sparsification/base_blockwise_sparsification.py -------------------------------------------------------------------------------- /llmc/compression/sparsification/dense.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/sparsification/dense.py -------------------------------------------------------------------------------- /llmc/compression/sparsification/kvsparse.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/sparsification/kvsparse.py -------------------------------------------------------------------------------- /llmc/compression/sparsification/magnitude.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/sparsification/magnitude.py -------------------------------------------------------------------------------- /llmc/compression/sparsification/shortgpt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/sparsification/shortgpt.py -------------------------------------------------------------------------------- /llmc/compression/sparsification/wanda.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/sparsification/wanda.py -------------------------------------------------------------------------------- /llmc/compression/token_reduction/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/token_reduction/__init__.py -------------------------------------------------------------------------------- /llmc/compression/token_reduction/base_blockwise_token_reduction.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/token_reduction/base_blockwise_token_reduction.py -------------------------------------------------------------------------------- /llmc/compression/token_reduction/dart.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/token_reduction/dart.py -------------------------------------------------------------------------------- /llmc/compression/token_reduction/divprune.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/token_reduction/divprune.py -------------------------------------------------------------------------------- /llmc/compression/token_reduction/dycoke.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/token_reduction/dycoke.py -------------------------------------------------------------------------------- /llmc/compression/token_reduction/fastervlm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/token_reduction/fastervlm.py -------------------------------------------------------------------------------- /llmc/compression/token_reduction/fastv.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/token_reduction/fastv.py -------------------------------------------------------------------------------- /llmc/compression/token_reduction/fastvid.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/token_reduction/fastvid.py -------------------------------------------------------------------------------- /llmc/compression/token_reduction/holitom.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/token_reduction/holitom.py -------------------------------------------------------------------------------- /llmc/compression/token_reduction/mustdrop.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/token_reduction/mustdrop.py -------------------------------------------------------------------------------- /llmc/compression/token_reduction/prunevid.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/token_reduction/prunevid.py -------------------------------------------------------------------------------- /llmc/compression/token_reduction/pyramiddrop.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/token_reduction/pyramiddrop.py -------------------------------------------------------------------------------- /llmc/compression/token_reduction/random.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/token_reduction/random.py -------------------------------------------------------------------------------- /llmc/compression/token_reduction/sparsevlm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/token_reduction/sparsevlm.py -------------------------------------------------------------------------------- /llmc/compression/token_reduction/token_reduction_module.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/token_reduction/token_reduction_module.py -------------------------------------------------------------------------------- /llmc/compression/token_reduction/tome.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/token_reduction/tome.py -------------------------------------------------------------------------------- /llmc/compression/token_reduction/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/token_reduction/utils.py -------------------------------------------------------------------------------- /llmc/compression/token_reduction/visionzip.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/token_reduction/visionzip.py -------------------------------------------------------------------------------- /llmc/compression/token_reduction/vispruner.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/token_reduction/vispruner.py -------------------------------------------------------------------------------- /llmc/compression/token_reduction/visualizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/compression/token_reduction/visualizer.py -------------------------------------------------------------------------------- /llmc/data/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/data/__init__.py -------------------------------------------------------------------------------- /llmc/data/dataset/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/data/dataset/__init__.py -------------------------------------------------------------------------------- /llmc/data/dataset/base_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/data/dataset/base_dataset.py -------------------------------------------------------------------------------- /llmc/data/dataset/specified_preproc.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/data/dataset/specified_preproc.py -------------------------------------------------------------------------------- /llmc/eval/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/eval/__init__.py -------------------------------------------------------------------------------- /llmc/eval/eval_acc.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/eval/eval_acc.py -------------------------------------------------------------------------------- /llmc/eval/eval_base.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/eval/eval_base.py -------------------------------------------------------------------------------- /llmc/eval/eval_code.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/eval/eval_code.py -------------------------------------------------------------------------------- /llmc/eval/eval_custom_generate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/eval/eval_custom_generate.py -------------------------------------------------------------------------------- /llmc/eval/eval_custom_generate_just_infer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/eval/eval_custom_generate_just_infer.py -------------------------------------------------------------------------------- /llmc/eval/eval_ppl.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/eval/eval_ppl.py -------------------------------------------------------------------------------- /llmc/eval/eval_token_consist.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/eval/eval_token_consist.py -------------------------------------------------------------------------------- /llmc/eval/eval_video_generate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/eval/eval_video_generate.py -------------------------------------------------------------------------------- /llmc/eval/eval_vqa.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/eval/eval_vqa.py -------------------------------------------------------------------------------- /llmc/eval/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/eval/utils.py -------------------------------------------------------------------------------- /llmc/models/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/__init__.py -------------------------------------------------------------------------------- /llmc/models/base_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/base_model.py -------------------------------------------------------------------------------- /llmc/models/bloom.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/bloom.py -------------------------------------------------------------------------------- /llmc/models/chatglm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/chatglm.py -------------------------------------------------------------------------------- /llmc/models/deepseekv2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/deepseekv2.py -------------------------------------------------------------------------------- /llmc/models/deepseekv3.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/deepseekv3.py -------------------------------------------------------------------------------- /llmc/models/falcon.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/falcon.py -------------------------------------------------------------------------------- /llmc/models/gemma2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/gemma2.py -------------------------------------------------------------------------------- /llmc/models/glm4v.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/glm4v.py -------------------------------------------------------------------------------- /llmc/models/internlm2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/internlm2.py -------------------------------------------------------------------------------- /llmc/models/internomni.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/internomni.py -------------------------------------------------------------------------------- /llmc/models/internvl2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/internvl2.py -------------------------------------------------------------------------------- /llmc/models/internvl3_5.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/internvl3_5.py -------------------------------------------------------------------------------- /llmc/models/llama.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/llama.py -------------------------------------------------------------------------------- /llmc/models/llava.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/llava.py -------------------------------------------------------------------------------- /llmc/models/llava_hf.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/llava_hf.py -------------------------------------------------------------------------------- /llmc/models/llava_onevision.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/llava_onevision.py -------------------------------------------------------------------------------- /llmc/models/minicpm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/minicpm.py -------------------------------------------------------------------------------- /llmc/models/minicpmv.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/minicpmv.py -------------------------------------------------------------------------------- /llmc/models/mistral.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/mistral.py -------------------------------------------------------------------------------- /llmc/models/mixtral.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/mixtral.py -------------------------------------------------------------------------------- /llmc/models/mllama.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/mllama.py -------------------------------------------------------------------------------- /llmc/models/opt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/opt.py -------------------------------------------------------------------------------- /llmc/models/phi.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/phi.py -------------------------------------------------------------------------------- /llmc/models/phi3.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/phi3.py -------------------------------------------------------------------------------- /llmc/models/qwen.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/qwen.py -------------------------------------------------------------------------------- /llmc/models/qwen2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/qwen2.py -------------------------------------------------------------------------------- /llmc/models/qwen2_5vl.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/qwen2_5vl.py -------------------------------------------------------------------------------- /llmc/models/qwen2audio.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/qwen2audio.py -------------------------------------------------------------------------------- /llmc/models/qwen2moe.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/qwen2moe.py -------------------------------------------------------------------------------- /llmc/models/qwen2vl.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/qwen2vl.py -------------------------------------------------------------------------------- /llmc/models/qwen3.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/qwen3.py -------------------------------------------------------------------------------- /llmc/models/qwen3moe.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/qwen3moe.py -------------------------------------------------------------------------------- /llmc/models/smollm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/smollm.py -------------------------------------------------------------------------------- /llmc/models/stablelm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/stablelm.py -------------------------------------------------------------------------------- /llmc/models/starcoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/starcoder.py -------------------------------------------------------------------------------- /llmc/models/videollava.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/videollava.py -------------------------------------------------------------------------------- /llmc/models/vila.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/vila.py -------------------------------------------------------------------------------- /llmc/models/vit.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/vit.py -------------------------------------------------------------------------------- /llmc/models/wan_i2v.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/wan_i2v.py -------------------------------------------------------------------------------- /llmc/models/wan_t2v.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/models/wan_t2v.py -------------------------------------------------------------------------------- /llmc/utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/utils/__init__.py -------------------------------------------------------------------------------- /llmc/utils/export_autoawq.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/utils/export_autoawq.py -------------------------------------------------------------------------------- /llmc/utils/export_lightx2v.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/utils/export_lightx2v.py -------------------------------------------------------------------------------- /llmc/utils/export_trtllm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/utils/export_trtllm.py -------------------------------------------------------------------------------- /llmc/utils/export_vllm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/utils/export_vllm.py -------------------------------------------------------------------------------- /llmc/utils/registry_factory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/utils/registry_factory.py -------------------------------------------------------------------------------- /llmc/utils/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/utils/utils.py -------------------------------------------------------------------------------- /llmc/utils/visualizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/llmc/utils/visualizer.py -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- 1 | -r requirements/runtime.txt 2 | -------------------------------------------------------------------------------- /requirements/docs.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/requirements/docs.txt -------------------------------------------------------------------------------- /requirements/runtime.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/requirements/runtime.txt -------------------------------------------------------------------------------- /scripts/run_llmc.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/scripts/run_llmc.sh -------------------------------------------------------------------------------- /scripts/run_lm_eval.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/scripts/run_lm_eval.sh -------------------------------------------------------------------------------- /tools/attention_vis.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/tools/attention_vis.py -------------------------------------------------------------------------------- /tools/download_calib_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/tools/download_calib_dataset.py -------------------------------------------------------------------------------- /tools/download_eval_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/tools/download_eval_dataset.py -------------------------------------------------------------------------------- /tools/llm_eval.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/tools/llm_eval.py -------------------------------------------------------------------------------- /tools/print_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/tools/print_model.py -------------------------------------------------------------------------------- /tools/quant_analysis.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ModelTC/LightCompress/HEAD/tools/quant_analysis.py --------------------------------------------------------------------------------