├── .gitignore ├── .pre-commit-config.yaml ├── LICENSE ├── README.md ├── assets ├── combine.jpg ├── open_compass_1223.jpg ├── performance_valley2.jpeg ├── performance_valley2_5.png ├── tts_inhouse_benchmark_1223.jpg ├── valley2_5_logo.png ├── valley2_5_structure.png ├── valley2_structure.png └── valley_logo.jpg ├── docs └── Valley2_5_Tech_Report.pdf ├── examples ├── demos_valley2 │ ├── demo_multi_image.py │ ├── demo_single_image.py │ └── demo_video.py └── demos_valley2_5 │ ├── demo_multi_image.py │ ├── demo_single_image.py │ └── demo_video.py ├── requirements.txt └── valley2 ├── constants.py ├── conversation.py ├── model ├── language_model │ ├── valley_qwen2.py │ └── valley_qwen2vl.py ├── multimodal_encoder │ ├── builder.py │ └── siglip_encoder.py ├── multimodal_projector │ └── builder.py ├── token_compressor │ ├── avgpool.py │ ├── builder.py │ ├── evo.py │ ├── lavit.py │ ├── minicpm_resampler.py │ └── roipool.py └── valley_arch.py ├── util ├── config.py ├── data_util.py ├── mm_utils.py └── vision_encoder_config.py ├── valley2_chat.py └── valley_utils.py /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/.gitignore -------------------------------------------------------------------------------- /.pre-commit-config.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/.pre-commit-config.yaml -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/LICENSE -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/README.md -------------------------------------------------------------------------------- /assets/combine.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/assets/combine.jpg -------------------------------------------------------------------------------- /assets/open_compass_1223.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/assets/open_compass_1223.jpg -------------------------------------------------------------------------------- /assets/performance_valley2.jpeg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/assets/performance_valley2.jpeg -------------------------------------------------------------------------------- /assets/performance_valley2_5.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/assets/performance_valley2_5.png -------------------------------------------------------------------------------- /assets/tts_inhouse_benchmark_1223.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/assets/tts_inhouse_benchmark_1223.jpg -------------------------------------------------------------------------------- /assets/valley2_5_logo.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/assets/valley2_5_logo.png -------------------------------------------------------------------------------- /assets/valley2_5_structure.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/assets/valley2_5_structure.png -------------------------------------------------------------------------------- /assets/valley2_structure.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/assets/valley2_structure.png -------------------------------------------------------------------------------- /assets/valley_logo.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/assets/valley_logo.jpg -------------------------------------------------------------------------------- /docs/Valley2_5_Tech_Report.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/docs/Valley2_5_Tech_Report.pdf -------------------------------------------------------------------------------- /examples/demos_valley2/demo_multi_image.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/examples/demos_valley2/demo_multi_image.py -------------------------------------------------------------------------------- /examples/demos_valley2/demo_single_image.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/examples/demos_valley2/demo_single_image.py -------------------------------------------------------------------------------- /examples/demos_valley2/demo_video.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/examples/demos_valley2/demo_video.py -------------------------------------------------------------------------------- /examples/demos_valley2_5/demo_multi_image.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/examples/demos_valley2_5/demo_multi_image.py -------------------------------------------------------------------------------- /examples/demos_valley2_5/demo_single_image.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/examples/demos_valley2_5/demo_single_image.py -------------------------------------------------------------------------------- /examples/demos_valley2_5/demo_video.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/examples/demos_valley2_5/demo_video.py -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/requirements.txt -------------------------------------------------------------------------------- /valley2/constants.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/valley2/constants.py -------------------------------------------------------------------------------- /valley2/conversation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/valley2/conversation.py -------------------------------------------------------------------------------- /valley2/model/language_model/valley_qwen2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/valley2/model/language_model/valley_qwen2.py -------------------------------------------------------------------------------- /valley2/model/language_model/valley_qwen2vl.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/valley2/model/language_model/valley_qwen2vl.py -------------------------------------------------------------------------------- /valley2/model/multimodal_encoder/builder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/valley2/model/multimodal_encoder/builder.py -------------------------------------------------------------------------------- /valley2/model/multimodal_encoder/siglip_encoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/valley2/model/multimodal_encoder/siglip_encoder.py -------------------------------------------------------------------------------- /valley2/model/multimodal_projector/builder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/valley2/model/multimodal_projector/builder.py -------------------------------------------------------------------------------- /valley2/model/token_compressor/avgpool.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/valley2/model/token_compressor/avgpool.py -------------------------------------------------------------------------------- /valley2/model/token_compressor/builder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/valley2/model/token_compressor/builder.py -------------------------------------------------------------------------------- /valley2/model/token_compressor/evo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/valley2/model/token_compressor/evo.py -------------------------------------------------------------------------------- /valley2/model/token_compressor/lavit.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/valley2/model/token_compressor/lavit.py -------------------------------------------------------------------------------- /valley2/model/token_compressor/minicpm_resampler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/valley2/model/token_compressor/minicpm_resampler.py -------------------------------------------------------------------------------- /valley2/model/token_compressor/roipool.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/valley2/model/token_compressor/roipool.py -------------------------------------------------------------------------------- /valley2/model/valley_arch.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/valley2/model/valley_arch.py -------------------------------------------------------------------------------- /valley2/util/config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/valley2/util/config.py -------------------------------------------------------------------------------- /valley2/util/data_util.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/valley2/util/data_util.py -------------------------------------------------------------------------------- /valley2/util/mm_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/valley2/util/mm_utils.py -------------------------------------------------------------------------------- /valley2/util/vision_encoder_config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/valley2/util/vision_encoder_config.py -------------------------------------------------------------------------------- /valley2/valley2_chat.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/valley2/valley2_chat.py -------------------------------------------------------------------------------- /valley2/valley_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/bytedance/Valley/HEAD/valley2/valley_utils.py --------------------------------------------------------------------------------