OneVision-Encoder
Codec-Aligned Sparsity as a Foundational Principle for Multimodal Intelligence
File Explorer
Download Latest Version (.zip)- issue-explainer.md
- logo_dark.png
- logo_light.png
- method.jpg
- result_ap.png
- hevc
- hevcdsp.h
- hevcpred.h
- openHevcWrapper.h
- __init__.py
- ap_dataloader_dali.py
- ap_dataloader_dali_codec.py
- data_decord_codec.py
- data_v2.py
- data_v2_multi_res.py
- data_v2_ocr.py
- data_v2_parallel_rec.py
- hevc_feature_decoder.py
- hevc_feature_decoder_mv.py
- __init__.py
- dataset_onevision_encoder.py
- generate_rec.py
- properties.py
- registry.py
- issue-105.html
- issue-112.html
- issue-113.html
- issue-116.html
- data_card.md
- index.html
- model_card.md
- common.sh
- common_codec.sh
- eval_aimv2_large_patch14_224_16frames.sh
- eval_aimv2_large_patch14_224_8frames.sh
- eval_clip_base_16frames.sh
- eval_clip_base_8frames.sh
- eval_clip_large_16frames.sh
- eval_clip_large_8frames.sh
- eval_dinov2_base_16frames.sh
- eval_dinov2_base_8frames.sh
- eval_dinov2_large_16frames.sh
- eval_dinov3_base_16frames.sh
- eval_dinov3_base_8frames.sh
- eval_dinov3_large_16frames.sh
- eval_dinov3_large_8frames.sh
- eval_metaclip2_large_16frames.sh
- eval_metaclip2_large_8frames.sh
- eval_metaclip_base_16frames.sh
- eval_metaclip_base_8frames.sh
- eval_metaclip_large_16frames.sh
- eval_metaclip_large_8frames.sh
- eval_ov_encoder_large_16frames.sh
- eval_ov_encoder_large_2kpatches_codec.sh
- eval_ov_encoder_large_4kpatches_codec.sh
- eval_ov_encoder_large_8frames.sh
- eval_siglip2_base_16frames.sh
- eval_siglip2_base_8frames.sh
- eval_siglip2_large_patch16_256_16frames.sh
- eval_siglip2_large_patch16_256_1frames.sh
- eval_siglip2_large_patch16_256_8frames.sh
- eval_siglip2_naflex_large_patch16_256_16frames.sh
- eval_siglip_base_16frames.sh
- eval_siglip_base_8frames.sh
- eval_siglip_large_patch16_256_16frames.sh
- eval_siglip_large_patch16_256_8frames.sh
- attentive_probe.py
- attentive_probe_codec.py
- h264idct_template.c
- hevc_refs.c
- hevcdec.c
- hevcdec.h
- hevcdsp_template.cpp
- patch.sh
- yuv2rgb.c
- install_ffmpeg.sh
- __init__.py
- __main__.py
- h264_api.cpp
- reader.py
- offline_precompute_llava_codec_assets.py
- stage1.py
- stage2.py
- dockerfile
- install.sh
- pyproject.toml
- README.md
- setup.py
- sample_XwB-dkS5BVM__ytb_XwB-dkS5BVM__58574aa9__caea9e43_T64_square576_p16_K10368.frame_ids.npy
- sample_XwB-dkS5BVM__ytb_XwB-dkS5BVM__58574aa9__caea9e43_T64_square576_p16_K10368.meta.json
- sample_XwB-dkS5BVM__ytb_XwB-dkS5BVM__58574aa9__caea9e43_T64_square576_p16_K10368.visidx_thw.npy
- sample_YVQwAEKZpaU__ytb_YVQwAEKZpaU__22bd1705__9c865b10_T64_square576_p16_K10368.frame_ids.npy
- sample_YVQwAEKZpaU__ytb_YVQwAEKZpaU__22bd1705__9c865b10_T64_square576_p16_K10368.meta.json
- sample_YVQwAEKZpaU__ytb_YVQwAEKZpaU__22bd1705__9c865b10_T64_square576_p16_K10368.visidx_thw.npy
- stage1_failed.shard00of01.txt
- stage1_unsupported.shard00of01.txt
- positions_thw.npy
- ytb_XwB-dkS5BVM_000.jpg
- ytb_XwB-dkS5BVM_001.jpg
- ytb_XwB-dkS5BVM_002.jpg
- ytb_XwB-dkS5BVM_003.jpg
- ytb_XwB-dkS5BVM_004.jpg
- ytb_XwB-dkS5BVM_005.jpg
- ytb_XwB-dkS5BVM_006.jpg
- ytb_XwB-dkS5BVM_007.jpg
- positions_thw.npy
- ytb_YVQwAEKZpaU_000.jpg
- ytb_YVQwAEKZpaU_001.jpg
- ytb_YVQwAEKZpaU_002.jpg
- ytb_YVQwAEKZpaU_003.jpg
- ytb_YVQwAEKZpaU_004.jpg
- ytb_YVQwAEKZpaU_005.jpg
- ytb_YVQwAEKZpaU_006.jpg
- ytb_YVQwAEKZpaU_007.jpg
- training_codec.jsonl
- training_codec.jsonl.fail.txt
- ytb_XwB-dkS5BVM.mp4
- ytb_YVQwAEKZpaU.mp4
- raw_video_samples.json
- run_training_data_pipeline.sh
- llava_qwen.py
- llava_qwen3.py
- modeling_llama.py
- builder.py
- onevision_encoder.py
- siglip2_naflex.py
- builder.py
- example_spatial_merge.py
- pooler_projector.py
- TRAINING_FLOW.md
- USAGE_SPATIAL_MERGE.md
- builder.py
- masked_drop.py
- perceiver.py
- qformer.py
- spatial_pool.py
- __init__.py
- apply_delta.py
- builder.py
- consolidate.py
- llava_arch.py
- make_delta.py
- utils.py
- llama_flash_attn_monkey_patch.py
- llava_trainer.py
- llava_trainer_eval.py
- train.py
- train_mem.py
- __init__.py
- constants.py
- conversation.py
- mm_utils.py
- utils.py
- claude.yml
- lint.yml
- issue_template.md
- pull_request_template.md
- README.md
- run_evalcicd.sh
- README_ar.md
- README_de.md
- README_es.md
- README_fr.md
- README_hi.md
- README_id.md
- README_it.md
- README_ja.md
- README_ko.md
- README_nl.md
- README_pl.md
- README_pt-BR.md
- README_ru.md
- README_tr.md
- README_vi.md
- README_zh-CN.md
- README_zh-TW.md
- caching.md
- commands.md
- current_tasks.md
- lmms-eval-0.3.md
- lmms-eval-0.4.md
- lmms-eval-0.5.md
- lmms-eval-0.6.md
- model_guide.md
- README.md
- run_examples.md
- task_guide.md
- throughput_metrics.md
- tool_call_qwen2_5_vl.jinja
- crop_video_mcp_server.py
- sample_mcp_server.py
- aero_1_audio.sh
- aria.sh
- async_openai_compatible.sh
- auroracap.sh
- bagel.sh
- cambrians.sh
- claude.sh
- gemma3.sh
- idefics2.sh
- instructblip.sh
- internvl1.5.sh
- internvl2.sh
- llama_vid.sh
- llama_vision.sh
- llava_1_5.sh
- llava_next.sh
- llava_onevision.sh
- llava_onevision1_5.sh
- llava_video.sh
- movie_chat.sh
- mplug_owl.sh
- openai_compatible.sh
- openrouter_molmo.sh
- plm.sh
- qwen25vl.sh
- qwen2vl.sh
- qwen3vl.sh
- sglang.sh
- sglang_qwen3vl.sh
- slime.sh
- tensor_parallel.sh
- thyme.sh
- tinyllava.sh
- video_chatgpt.sh
- video_llava.sh
- vllm_generate_qwen3vl.sh
- vllm_qwen2vl.sh
- vllm_qwen3vl.sh
- xai_grok.sh
- xcomposer.sh
- __init__.py
- filter.py
- group.py
- instance.py
- metrics.py
- model.py
- registry.py
- samplers.py
- task.py
- __init__.py
- loader.py
- registry.py
- __init__.py
- cache.py
- __init__.py
- client.py
- http_server.py
- job_scheduler.py
- protocol.py
- server_args.py
- __init__.py
- decontamination.py
- extraction.py
- selection.py
- transformation.py
- __init__.py
- base.py
- sglang.py
- __init__.py
- async_azure_openai.py
- async_openai.py
- azure_openai.py
- dummy.py
- openai.py
- __init__.py
- base.py
- factory.py
- prompt.py
- protocol.py
- utils.py
- __init__.py
- evaluation_tracker.py
- utils.py
- wandb_logger.py
- __init__.py
- client.py
- async_openai.py
- bagel_lmms_engine.py
- huggingface.py
- llava_hf.py
- llava_onevision1_5.py
- longvila.py
- openai_compatible.py
- qwen2_5_vl.py
- qwen3_vl.py
- sglang.py
- thyme.py
- vllm.py
- vllm_generate.py
- qwen_generate_utils.py
- sandbox.py
- utils.py
- __init__.py
- audio_processing.py
- gen_metrics.py
- load_video.py
- reasoning_model_utils.py
- aero.py
- aria.py
- auroracap.py
- bagel.py
- batch_gpt4.py
- cambrian.py
- cambrians.py
- claude.py
- cogvlm2.py
- egogpt.py
- from_log.py
- fuyu.py
- gemini_api.py
- gemma3.py
- gpt4o_audio.py
- gpt4v.py
- idefics2.py
- instructblip.py
- internvideo2.py
- internvideo2_5.py
- internvl.py
- internvl2.py
- internvl3.py
- internvl3_5.py
- llama_vid.py
- llama_vision.py
- llava.py
- llava_hf.py
- llava_onevision.py
- llava_onevision1_5.py
- llava_onevision_moviechat.py
- llava_ov_encoder.py
- llava_sglang.py
- llava_vid.py
- longva.py
- mantis.py
- minicpm_v.py
- minimonkey.py
- moviechat.py
- mplug_owl_video.py
- ola.py
- openai_compatible.py
- oryx.py
- phi3v.py
- phi4_multimodal.py
- plm.py
- qwen2_5_omni.py
- qwen2_5_vl.py
- qwen2_5_vl_interleave.py
- qwen2_audio.py
- qwen2_vl.py
- qwen3_omni.py
- qwen3_vl.py
- qwen_vl.py
- qwen_vl_api.py
- reka.py
- ross.py
- slime.py
- srt_api.py
- tinyllava.py
- video_chatgpt.py
- video_llava.py
- video_salmonn_2.py
- videochat2.py
- videochat_flash.py
- videollama3.py
- vila.py
- vita.py
- vllm.py
- vora.py
- whisper.py
- whisper_vllm.py
- xcomposer2_4KHD.py
- xcomposer2d5.py
- __init__.py
- whisper_tt.py
- file_utils.py
- gpt_eval_utils.py
- math_verify_utils.py
- reasoning_utils.py
- video_loader.py
- vqa_eval_metric.py
- _default_template_yaml
- activitynetqa_generation.yaml
- utils.py
- ai2d_reasoning.yaml
- utils.py
- ai2d.yaml
- ai2d_lite.yaml
- ai2d_no_mask.yaml
- upload_ai2d.py
- utils.py
- aime24_nofigures_agg8.yaml
- aime25_nofigures_agg8.yaml
- utils.py
- aime24_figures.yaml
- aime24_figures_agg64.yaml
- aime24_nofigures.yaml
- aime24_nofigures_agg64.yaml
- aime24_nofigures_agg8.yaml
- aime25_nofigures.yaml
- aime25_nofigures_agg64.yaml
- aime25_nofigures_agg8.yaml
- aime_2024_agg8.yaml
- aime_2024_rebase.yaml
- aime_figures.yaml
- aime_nofigures.yaml
- README.md
- utils.py
- _default_template_yaml
- air_bench_chat.yaml
- air_bench_chat_mixed.yaml
- air_bench_chat_music.yaml
- air_bench_chat_sound.yaml
- air_bench_chat_speech.yaml
- air_bench_foundation.yaml
- air_bench_foundation_music.yaml
- air_bench_foundation_sound.yaml
- air_bench_foundation_speech.yaml
- utils.py
- alpaca_audio.yaml
- utils.py
- amber_g.yaml
- utils.py
- arc_challenge.yaml
- arc_easy.yaml
- README.md
- av_odyssey.yaml
- utils.py
- av_speakerbench_audio.yaml
- av_speakerbench_audiovisual.yaml
- av_speakerbench_visual.yaml
- utils.py
- __init__.py
- babyvision.yaml
- prompt.py
- README.md
- utils.py
- __init__.py
- babyvision_gen.yaml
- prompt.py
- utils.py
- _default_template_yaml
- blink.yaml
- blink_art_style.yaml
- blink_counting.yaml
- blink_forensic_detection.yaml
- blink_functional_correspondence.yaml
- blink_iq_test.yaml
- blink_jigsaw.yaml
- blink_multi_view_reasoning.yaml
- blink_object_localization.yaml
- blink_relative_depth.yaml
- blink_relative_reflectance.yaml
- blink_semantic_correspondence.yaml
- blink_spatial_relation.yaml
- blink_visual_correspondence.yaml
- blink_visual_similarity.yaml
- utils.py
- camerabench_vqa.yaml
- utils.py
- _default_template_yaml
- capability.yaml
- capability_action.yaml
- capability_camera_angle.yaml
- capability_camera_movement.yaml
- capability_character_identification.yaml
- capability_dynamic_object_number.yaml
- capability_event.yaml
- capability_object_category.yaml
- capability_object_color.yaml
- capability_object_number.yaml
- capability_OCR.yaml
- capability_scene.yaml
- capability_spatial_relation.yaml
- capability_style.yaml
- prompt.py
- utils.py
- _default_template_yaml
- captionqa.yaml
- captionqa_document.yaml
- captionqa_ecommerce.yaml
- captionqa_embodiedai.yaml
- captionqa_natural.yaml
- README.md
- utils.py
- charades.yaml
- eval_tvg.py
- utils.py
- chartqa.yaml
- chartqa_lite.yaml
- upload_chartqa.py
- utils.py
- _default_template_yaml
- charxiv.yaml
- charxiv_val_descriptive.yaml
- charxiv_val_reasoning.yaml
- utils.py
- _default_template_yaml
- charxiv.yaml
- charxiv_val_descriptive.yaml
- charxiv_val_reasoning.yaml
- constant.py
- descriptive_utils.py
- reasoning_utils.py
- utils.py
- cinepile.yaml
- utils.py
- _default_template_yaml
- clotho_aqa.yaml
- clotho_aqa_test.yaml
- clotho_aqa_val.yaml
- clotho_asqa_test_v2.yaml
- utils.py
- _cmmmu.yaml
- _default_template_cmmmu_yaml
- cmmmu_test.yaml
- cmmmu_val.yaml
- utils.py
- coco2014_cap.yaml
- coco2014_cap_test.yaml
- coco2014_cap_val.yaml
- coco2017_cap.yaml
- coco2017_cap_test.yaml
- coco2017_cap_val.yaml
- coco2017_cap_val_lite.yaml
- coco_cap.yaml
- coco_karpathy.yaml
- coco_karpathy_test.yaml
- coco_karpathy_val.yaml
- utils.py
- coco_cap_chair.yaml
- utils.py
- _default_template_yaml
- common_voice_15.yaml
- common_voice_15_en.yaml
- common_voice_15_fr.yaml
- common_voice_15_zh-CN.yaml
- utils.py
- conbench.yaml
- utils.py
- _default_template_en_zh_yaml
- _default_template_zh_en_yaml
- covost2.yaml
- covost2_en_zh.yaml
- covost2_en_zh_dev.yaml
- covost2_en_zh_test.yaml
- covost2_zh_en.yaml
- covost2_zh_en_dev.yaml
- covost2_zh_en_test.yaml
- utils.py
- csbench.yaml
- csbench_assertion.yaml
- csbench_mcq.yaml
- utils.py
- _default_template_yaml
- cuva.yaml
- cuva_test.yaml
- utils.py
- _default_template_yaml
- cv_bench.yaml
- cv_bench_2d.yaml
- cv_bench_3d.yaml
- utils.py
- _cvrr.yaml
- _default_template_yaml
- cvrr_fine_grained_action_understanding.yaml
- cvrr_interpretation_of_social_context.yaml
- cvrr_interpretation_of_visual_context.yaml
- cvrr_multiple_actions_in_a_single_video.yaml
- cvrr_non_existent_actions_with_existent_scene_depictions.yaml
- cvrr_non_existent_actions_with_non_existent_scene_depictions.yaml
- cvrr_object_instance_count.yaml
- cvrr_partial_actions.yaml
- cvrr_time_order_understanding.yaml
- cvrr_understanding_emotional_context.yaml
- cvrr_unusual_and_physically_anomalous_activities.yaml
- utils.py
- _default_template_detailcaps_yaml
- detailcaps.yaml
- utils.py
- _default_template_docvqa_yaml
- docvqa.yaml
- docvqa_test.yaml
- docvqa_val.yaml
- docvqa_val_lite.yaml
- utils.py
- dtcbench.yaml
- utils.py
- _default_template_yaml
- _generate_config.py
- dynamath.yaml
- utils.py
- egoplan.yaml
- utils.py
- _default_template_yaml
- egoschema.yaml
- egoschema_mcppl.yaml
- egoschema_subset.yaml
- egoschema_subset_mcppl.yaml
- README.md
- utils.py
- _default_template_yaml
- egothink.yaml
- egothink_activity.yaml
- egothink_affordance.yaml
- egothink_assistance.yaml
- egothink_attribute.yaml
- egothink_comparing.yaml
- egothink_counting.yaml
- egothink_existence.yaml
- egothink_forecasting.yaml
- egothink_location.yaml
- egothink_navigation.yaml
- egothink_situated.yaml
- egothink_spatial.yaml
- utils.py
- _default_template_yaml
- embspatial.yaml
- utils.py
- emma_all.yaml
- emma_mini_all.yaml
- utils.py
- _default_template_yaml
- erqa.yaml
- utils.py
- _default_template_yaml
- FALCONBench_mcq.yaml
- FALCONBench_mcq_temploc.yaml
- FALCONBench_oq.yaml
- FALCONBench_oq_temploc.yaml
- README.md
- utils.py
- ferret.yaml
- rule.json
- utils.py
- _default_template_yaml
- fleurs.yaml
- fleurs_cmn_hans_cn.yaml
- fleurs_en.yaml
- fleurs_yue_hant_hk.yaml
- utils.py
- flickr30k.yaml
- flickr30k_test.yaml
- flickr30k_test_lite.yaml
- utils.py
- _default_template_yaml
- funqa.yaml
- funqa_test.yaml
- utils.py
- __init__.py
- openai_backend.py
- parse_prompt.py
- utils.py
- vie_prompts.py
- __init__.py
- gedit_bench.yaml
- README.md
- utils.py
- basic.py
- english.json
- english.py
- gigaspeech.yaml
- gigaspeech_dev.yaml
- gigaspeech_l_dev.yaml
- gigaspeech_l_test.yaml
- gigaspeech_m_dev.yaml
- gigaspeech_m_test.yaml
- gigaspeech_s_dev.yaml
- gigaspeech_s_test.yaml
- gigaspeech_test.yaml
- gigaspeech_xl_dev.yaml
- gigaspeech_xl_test.yaml
- gigaspeech_xs_dev.yaml
- gigaspeech_xs_test.yaml
- utils.py
- _generate_configs.py
- _gpqa_cot_n_shot_yaml
- gpqa_diamond_cot_n_shot.yaml
- gpqa_extended_cot_n_shot.yaml
- gpqa_main_cot_n_shot.yaml
- utils.py
- _generate_configs.py
- _gpqa_cot_zeroshot_yaml
- gpqa_diamond_cot_zeroshot.yaml
- gpqa_extended_cot_zeroshot.yaml
- gpqa_main_cot_zeroshot.yaml
- utils.py
- _generate_configs.py
- _gpqa_generative_n_shot_yaml
- gpqa_diamond_generative_n_shot.yaml
- gpqa_extended_generative_n_shot.yaml
- gpqa_main_generative_n_shot.yaml
- utils.py
- _generate_configs.py
- _gpqa_n_shot_yaml
- gpqa_diamond_n_shot.yaml
- gpqa_extended_n_shot.yaml
- gpqa_main_n_shot.yaml
- utils.py
- gpqa_diamond_openai.yaml
- gpqa_diamond_openai_agg64.yaml
- gpqa_diamond_openai_maj64_cov64.yaml
- utils.py
- gpqa_thinking.yaml
- utils.py
- _generate_configs.py
- _gpqa_zeroshot_yaml
- gpqa_diamond_zeroshot.yaml
- gpqa_extended_zeroshot.yaml
- gpqa_main_zeroshot.yaml
- utils.py
- README.md
- gqa.yaml
- gqa_lite.yaml
- utils.py
- gqa_ru.yaml
- utils.py
- groundingme.yaml
- utils.py
- gsm8k-cot-llama.yaml
- gsm8k-cot-self-consistency.yaml
- gsm8k-cot-zeroshot.yaml
- gsm8k-cot.yaml
- gsm8k.yaml
- README.md
- evaluate_hb.py
- hallusion_bench_image.yaml
- utils.py
- hellaswag.yaml
- README.md
- utils.py
- hrbench.yaml
- hrbench4k.yaml
- hrbench8k.yaml
- hrbench_evals.py
- utils.py
- _default_template_docvqa_yaml
- iconqa.yaml
- iconqa_test.yaml
- iconqa_val.yaml
- utils.py
- ifeval.yaml
- instructions.py
- instructions_registry.py
- instructions_util.py
- README.md
- utils.py
- ii_bench.yaml
- utils.py
- illusionvqa.yaml
- illusionvqa_comprehension.yaml
- illusionvqa_soft_localization.yaml
- utils.py
- __init__.py
- imgedit.yaml
- prompt.py
- README.md
- utils.py
- _default_template_infovqa_yaml
- infovqa.yaml
- infovqa_test.yaml
- infovqa_val.yaml
- infovqa_val_lite.yaml
- utils.py
- _default_template_internal_eval_yaml
- d170_cn.yaml
- d170_cn_utils.py
- d170_en.yaml
- d170_en_utils.py
- dc100_en.yaml
- dc100_en_utils.py
- dc200_cn.yaml
- dc200_cn_utils.py
- internal_eval.yaml
- utils.py
- _default_template_yaml
- jmmmu.yaml
- jmmmu_accounting.yaml
- jmmmu_agriculture.yaml
- jmmmu_architecture_and_engineering.yaml
- jmmmu_basic_medical_science.yaml
- jmmmu_biology.yaml
- jmmmu_chemistry.yaml
- jmmmu_clinical_medicine.yaml
- jmmmu_computer_science.yaml
- jmmmu_design.yaml
- jmmmu_diagnostics_and_laboratory_medicine.yaml
- jmmmu_economics.yaml
- jmmmu_electronics.yaml
- jmmmu_energy_and_power.yaml
- jmmmu_finance.yaml
- jmmmu_japanese_art.yaml
- jmmmu_japanese_heritage.yaml
- jmmmu_japanese_history.yaml
- jmmmu_manage.yaml
- jmmmu_marketing.yaml
- jmmmu_materials.yaml
- jmmmu_math.yaml
- jmmmu_mechanical_engineering.yaml
- jmmmu_music.yaml
- jmmmu_pharmacy.yaml
- jmmmu_physics.yaml
- jmmmu_psychology.yaml
- jmmmu_public_health.yaml
- jmmmu_world_history.yaml
- utils.py
- _default_template_yaml
- jmmmu_pro.yaml
- jmmmu_pro_cot.yaml
- jmmmu_pro_standard.yaml
- jmmmu_pro_standard_cot.yaml
- jmmmu_pro_vision.yaml
- jmmmu_pro_vision_cot.yaml
- utils.py
- k12.yaml
- utils.py
- __init__.py
- kris_bench.yaml
- prompt.py
- README.md
- utils.py
- lemonade.yaml
- README.md
- utils.py
- basic.py
- english.json
- english.py
- _default_yaml_template
- cn_tn.py
- librispeech.yaml
- librispeech_dev_clean.yaml
- librispeech_dev_other.yaml
- librispeech_long.yaml
- librispeech_test_clean.yaml
- librispeech_test_clean_long.yaml
- librispeech_test_other.yaml
- librispeech_test_other_long.yaml
- utils.py
- live_bench.yaml
- live_bench_2406.yaml
- live_bench_2407.yaml
- live_bench_2409.yaml
- live_bench_template_yaml
- live_bench_template_yaml_v2
- utils.py
- utils_v2.py
- livexiv_tqa.yaml
- livexiv_tqa_template_yaml
- livexiv_tqa_v1.yaml
- livexiv_tqa_v2.yaml
- livexiv_tqa_v3.yaml
- livexiv_tqa_v4.yaml
- livexiv_tqa_v5.yaml
- livexiv_tqa_v6.yaml
- utils.py
- livexiv_vqa.yaml
- livexiv_vqa_template_yaml
- livexiv_vqa_v1.yaml
- livexiv_vqa_v2.yaml
- livexiv_vqa_v3.yaml
- livexiv_vqa_v4.yaml
- livexiv_vqa_v5.yaml
- livexiv_vqa_v6.yaml
- utils.py
- llava-bench-coco.yaml
- rule.json
- utils.py
- llava-in-the-wild.yaml
- llava-in-the-wild_ko.yaml
- rule.json
- rule_ko.json
- utils.py
- utils_ko.py
- _default_template_interleave_yaml
- in_domain.yaml
- interleave_bench.yaml
- multi_view_in_domain.yaml
- out_of_domain.yaml
- utils.py
- _default_template_wilder_yaml
- llava_wilder_small.yaml
- utils.py
- logicvista_thinking.yaml
- utils.py
- longtimescope.yaml
- utils.py
- longvideobench_no_visual.yaml
- utils.py
- longvideobench_random_choice.yaml
- utils.py
- longvideobench_test_i.yaml
- longvideobench_test_v.yaml
- longvideobench_val_i.yaml
- longvideobench_val_v.yaml
- utils.py
- longvt_no_visual.yaml
- utils.py
- longvt_non_think.yaml
- longvt_reasoning.yaml
- longvt_tool.yaml
- README.md
- utils.py
- lsdbench.yaml
- README.md
- utils.py
- lvbench_no_visual.yaml
- utils.py
- lvbench_random_choice.yaml
- utils.py
- lvbench.yaml
- utils.py
- mantis.yaml
- utils.py
- mathverse_testmini.yaml
- utils.py
- mathverse.yaml
- mathverse_evals.py
- mathverse_testmini.yaml
- mathverse_testmini_text.yaml
- mathverse_testmini_text_dominant.yaml
- mathverse_testmini_text_lite.yaml
- mathverse_testmini_text_only.yaml
- mathverse_testmini_vision.yaml
- mathverse_testmini_vision_dominant.yaml
- mathverse_testmini_vision_intensive.yaml
- mathverse_testmini_vision_only.yaml
- utils.py
- _default_template_yaml
- mathvision_reason_test.yaml
- mathvision_reason_testmini.yaml
- utils.py
- eval_utils.py
- mathvision_reason_test.yaml
- mathvision_reason_testmini.yaml
- mathvision_test.yaml
- mathvision_testmini.yaml
- utils.py
- mathvista_testmini_cot.yaml
- utils.py
- mathvista.yaml
- mathvista_evals.py
- mathvista_test.yaml
- mathvista_testmini.yaml
- mathvista_testmini_cot.yaml
- mathvista_testmini_format.yaml
- mathvista_testmini_solution.yaml
- utils.py
- medqa.yaml
- utils.py
- all_task_meta.json
- analysis_utils.py
- derive_breakdown_results.py
- mean_agg.py
- min_agg.py
- unsupported_agg.py
- parsers.py
- utils.py
- answer_str_parse.py
- dummy_parse.py
- json_parse.py
- conversions.py
- metrics.py
- transformations.py
- ascii_art_vlm_judge.py
- chess_jaccard.py
- constrained_generation.py
- coordinate_sequence_match.py
- dict_equality.py
- dict_exact_match_agg_recall.py
- dict_jaccard_agg_jaccard.py
- dict_nbbox_iou_tuple_agg_jaccard.py
- dict_set_equality_agg_jaccard.py
- exact_str_match.py
- exact_str_match_case_insensitive.py
- general_numerical_match.py
- geo_proximity.py
- gleu.py
- jaccard.py
- latex_expr_equality.py
- longest_common_list_prefix_ratio.py
- mse.py
- multi_ref_phrase.py
- nbbox_iou.py
- near_str_match.py
- nli_entailment.py
- normalized_similarity_damerau_levenshtein.py
- number_rel_diff_ratio.py
- positive_int_match.py
- program_judge.py
- sacrebleu_bleu.py
- sequence_equality.py
- set_equality.py
- set_precision.py
- simple_str_match.py
- symbolic_planning.py
- unsupported_scoring.py
- vlm_as_judge.py
- xml_nbbox_iou.py
- xml_norm_point_distance.py
- xml_norm_point_in_bbox.py
- __init__.py
- aggregation_type.py
- metric_type.py
- README.md
- response_parse_type.py
- _default_template_yaml
- evaluator.py
- image_video_utils.py
- megabench.yaml
- megabench_core.yaml
- megabench_core_si.yaml
- megabench_open.yaml
- megabench_open_si.yaml
- README.md
- utils.py
- mia_bench.yaml
- utils.py
- _default_template_yaml
- mindcube_full.yaml
- mindcube_full_among.yaml
- mindcube_full_around.yaml
- mindcube_full_rotation.yaml
- mindcube_tiny.yaml
- mindcube_tiny_among.yaml
- mindcube_tiny_around.yaml
- mindcube_tiny_rotation.yaml
- utils.py
- mirb.yaml
- utils.py
- _default_template_yaml
- mix_evals_audio2_text_freeform.yaml
- mix_evals_audio2_text_freeform_hard.yaml
- mix_evals_audio2text.yaml
- mix_evals_audio2text_hard.yaml
- utils.py
- _default_template_yaml
- mix_evals_image2text.yaml
- mix_evals_image2text_freeform.yaml
- mix_evals_image2text_freeform_hard.yaml
- mix_evals_image2text_hard.yaml
- mix_evals_image2text_mc.yaml
- mix_evals_image2text_mc_hard.yaml
- utils.py
- _default_template_yaml
- mix_evals_video2text.yaml
- mix_evals_video2text_freeform.yaml
- mix_evals_video2text_freeform_hard.yaml
- mix_evals_video2text_hard.yaml
- mix_evals_video2text_mc.yaml
- mix_evals_video2text_mc_hard.yaml
- mix_evals_video2text_openended.yaml
- utils.py
- README.md
- mlvu_dev.yaml
- mlvu_test.yaml
- utils.py
- _default_template_yaml
- mmau.yaml
- mmau_test.yaml
- mmau_test_mini.yaml
- utils.py
- _default_template_mmbench_en_reasoning_yaml
- mmbench_en_dev_reasoning.yaml
- mmbench_en_reasoning.yaml
- utils.py
- _default_template_mmbench_cn_yaml
- _default_template_mmbench_en_yaml
- _default_template_mmbench_ko_yaml
- _default_template_mmbench_ru_yaml
- cc_utils.py
- cn_utils.py
- en_utils.py
- ko_utils.py
- mmbench.yaml
- mmbench_cc.yaml
- mmbench_cn.yaml
- mmbench_cn_dev.yaml
- mmbench_cn_dev_lite.yaml
- mmbench_cn_test.yaml
- mmbench_en.yaml
- mmbench_en_dev.yaml
- mmbench_en_dev_lite.yaml
- mmbench_en_test.yaml
- mmbench_evals.py
- mmbench_ko_dev.yaml
- mmbench_ru_dev.yaml
- ru_utils.py
- mme.yaml
- utils.py
- mme_cot_direct.yaml
- mme_cot_reason.yaml
- README.md
- utils.py
- mme_realworld.yaml
- mme_realworld_cn.yaml
- mme_realworld_lite.yaml
- utils.py
- mme_sci.yaml
- run_eval.sh
- utils.py
- mme_sci_image.yaml
- run_eval.sh
- utils.py
- _continuation_template_yaml
- _mmlu.yaml
- mmlu_abstract_algebra.yaml
- mmlu_anatomy.yaml
- mmlu_astronomy.yaml
- mmlu_business_ethics.yaml
- mmlu_clinical_knowledge.yaml
- mmlu_college_biology.yaml
- mmlu_college_chemistry.yaml
- mmlu_college_computer_science.yaml
- mmlu_college_mathematics.yaml
- mmlu_college_medicine.yaml
- mmlu_college_physics.yaml
- mmlu_computer_security.yaml
- mmlu_conceptual_physics.yaml
- mmlu_econometrics.yaml
- mmlu_electrical_engineering.yaml
- mmlu_elementary_mathematics.yaml
- mmlu_formal_logic.yaml
- mmlu_global_facts.yaml
- mmlu_high_school_biology.yaml
- mmlu_high_school_chemistry.yaml
- mmlu_high_school_computer_science.yaml
- mmlu_high_school_european_history.yaml
- mmlu_high_school_geography.yaml
- mmlu_high_school_government_and_politics.yaml
- mmlu_high_school_macroeconomics.yaml
- mmlu_high_school_mathematics.yaml
- mmlu_high_school_microeconomics.yaml
- mmlu_high_school_physics.yaml
- mmlu_high_school_psychology.yaml
- mmlu_high_school_statistics.yaml
- mmlu_high_school_us_history.yaml
- mmlu_high_school_world_history.yaml
- mmlu_human_aging.yaml
- mmlu_human_sexuality.yaml
- mmlu_international_law.yaml
- mmlu_jurisprudence.yaml
- mmlu_logical_fallacies.yaml
- mmlu_machine_learning.yaml
- mmlu_management.yaml
- mmlu_marketing.yaml
- mmlu_medical_genetics.yaml
- mmlu_miscellaneous.yaml
- mmlu_moral_disputes.yaml
- mmlu_moral_scenarios.yaml
- mmlu_nutrition.yaml
- mmlu_philosophy.yaml
- mmlu_prehistory.yaml
- mmlu_professional_accounting.yaml
- mmlu_professional_law.yaml
- mmlu_professional_medicine.yaml
- mmlu_professional_psychology.yaml
- mmlu_public_relations.yaml
- mmlu_security_studies.yaml
- mmlu_sociology.yaml
- mmlu_us_foreign_policy.yaml
- mmlu_virology.yaml
- mmlu_world_religions.yaml
- _default_template_yaml
- _mmlu.yaml
- _mmlu_humanities.yaml
- _mmlu_other.yaml
- _mmlu_social_sciences.yaml
- _mmlu_stem.yaml
- mmlu_abstract_algebra.yaml
- mmlu_anatomy.yaml
- mmlu_astronomy.yaml
- mmlu_business_ethics.yaml
- mmlu_clinical_knowledge.yaml
- mmlu_college_biology.yaml
- mmlu_college_chemistry.yaml
- mmlu_college_computer_science.yaml
- mmlu_college_mathematics.yaml
- mmlu_college_medicine.yaml
- mmlu_college_physics.yaml
- mmlu_computer_security.yaml
- mmlu_conceptual_physics.yaml
- mmlu_econometrics.yaml
- mmlu_electrical_engineering.yaml
- mmlu_elementary_mathematics.yaml
- mmlu_formal_logic.yaml
- mmlu_global_facts.yaml
- mmlu_high_school_biology.yaml
- mmlu_high_school_chemistry.yaml
- mmlu_high_school_computer_science.yaml
- mmlu_high_school_european_history.yaml
- mmlu_high_school_geography.yaml
- mmlu_high_school_government_and_politics.yaml
- mmlu_high_school_macroeconomics.yaml
- mmlu_high_school_mathematics.yaml
- mmlu_high_school_microeconomics.yaml
- mmlu_high_school_physics.yaml
- mmlu_high_school_psychology.yaml
- mmlu_high_school_statistics.yaml
- mmlu_high_school_us_history.yaml
- mmlu_high_school_world_history.yaml
- mmlu_human_aging.yaml
- mmlu_human_sexuality.yaml
- mmlu_international_law.yaml
- mmlu_jurisprudence.yaml
- mmlu_logical_fallacies.yaml
- mmlu_machine_learning.yaml
- mmlu_management.yaml
- mmlu_marketing.yaml
- mmlu_medical_genetics.yaml
- mmlu_miscellaneous.yaml
- mmlu_moral_disputes.yaml
- mmlu_moral_scenarios.yaml
- mmlu_nutrition.yaml
- mmlu_philosophy.yaml
- mmlu_prehistory.yaml
- mmlu_professional_accounting.yaml
- mmlu_professional_law.yaml
- mmlu_professional_medicine.yaml
- mmlu_professional_psychology.yaml
- mmlu_public_relations.yaml
- mmlu_security_studies.yaml
- mmlu_sociology.yaml
- mmlu_us_foreign_policy.yaml
- mmlu_virology.yaml
- mmlu_world_religions.yaml
- _cot_prompts.json
- _mmlu.yaml
- _mmlu_flan_cot_fewshot_template_yaml
- mmlu_abstract_algebra.yaml
- mmlu_anatomy.yaml
- mmlu_astronomy.yaml
- mmlu_business_ethics.yaml
- mmlu_clinical_knowledge.yaml
- mmlu_college_biology.yaml
- mmlu_college_chemistry.yaml
- mmlu_college_computer_science.yaml
- mmlu_college_mathematics.yaml
- mmlu_college_medicine.yaml
- mmlu_college_physics.yaml
- mmlu_computer_security.yaml
- mmlu_conceptual_physics.yaml
- mmlu_econometrics.yaml
- mmlu_electrical_engineering.yaml
- mmlu_elementary_mathematics.yaml
- mmlu_formal_logic.yaml
- mmlu_global_facts.yaml
- mmlu_high_school_biology.yaml
- mmlu_high_school_chemistry.yaml
- mmlu_high_school_computer_science.yaml
- mmlu_high_school_european_history.yaml
- mmlu_high_school_geography.yaml
- mmlu_high_school_government_and_politics.yaml
- mmlu_high_school_macroeconomics.yaml
- mmlu_high_school_mathematics.yaml
- mmlu_high_school_microeconomics.yaml
- mmlu_high_school_physics.yaml
- mmlu_high_school_psychology.yaml
- mmlu_high_school_statistics.yaml
- mmlu_high_school_us_history.yaml
- mmlu_high_school_world_history.yaml
- mmlu_human_aging.yaml
- mmlu_human_sexuality.yaml
- mmlu_international_law.yaml
- mmlu_jurisprudence.yaml
- mmlu_logical_fallacies.yaml
- mmlu_machine_learning.yaml
- mmlu_management.yaml
- mmlu_marketing.yaml
- mmlu_medical_genetics.yaml
- mmlu_miscellaneous.yaml
- mmlu_moral_disputes.yaml
- mmlu_moral_scenarios.yaml
- mmlu_nutrition.yaml
- mmlu_philosophy.yaml
- mmlu_prehistory.yaml
- mmlu_professional_accounting.yaml
- mmlu_professional_law.yaml
- mmlu_professional_medicine.yaml
- mmlu_professional_psychology.yaml
- mmlu_public_relations.yaml
- mmlu_security_studies.yaml
- mmlu_sociology.yaml
- mmlu_us_foreign_policy.yaml
- mmlu_virology.yaml
- mmlu_world_religions.yaml
- _mmlu.yaml
- _mmlu_flan_cot_zeroshot_template_yaml
- mmlu_abstract_algebra.yaml
- mmlu_anatomy.yaml
- mmlu_astronomy.yaml
- mmlu_business_ethics.yaml
- mmlu_clinical_knowledge.yaml
- mmlu_college_biology.yaml
- mmlu_college_chemistry.yaml
- mmlu_college_computer_science.yaml
- mmlu_college_mathematics.yaml
- mmlu_college_medicine.yaml
- mmlu_college_physics.yaml
- mmlu_computer_security.yaml
- mmlu_conceptual_physics.yaml
- mmlu_econometrics.yaml
- mmlu_electrical_engineering.yaml
- mmlu_elementary_mathematics.yaml
- mmlu_formal_logic.yaml
- mmlu_global_facts.yaml
- mmlu_high_school_biology.yaml
- mmlu_high_school_chemistry.yaml
- mmlu_high_school_computer_science.yaml
- mmlu_high_school_european_history.yaml
- mmlu_high_school_geography.yaml
- mmlu_high_school_government_and_politics.yaml
- mmlu_high_school_macroeconomics.yaml
- mmlu_high_school_mathematics.yaml
- mmlu_high_school_microeconomics.yaml
- mmlu_high_school_physics.yaml
- mmlu_high_school_psychology.yaml
- mmlu_high_school_statistics.yaml
- mmlu_high_school_us_history.yaml
- mmlu_high_school_world_history.yaml
- mmlu_human_aging.yaml
- mmlu_human_sexuality.yaml
- mmlu_international_law.yaml
- mmlu_jurisprudence.yaml
- mmlu_logical_fallacies.yaml
- mmlu_machine_learning.yaml
- mmlu_management.yaml
- mmlu_marketing.yaml
- mmlu_medical_genetics.yaml
- mmlu_miscellaneous.yaml
- mmlu_moral_disputes.yaml
- mmlu_moral_scenarios.yaml
- mmlu_nutrition.yaml
- mmlu_philosophy.yaml
- mmlu_prehistory.yaml
- mmlu_professional_accounting.yaml
- mmlu_professional_law.yaml
- mmlu_professional_medicine.yaml
- mmlu_professional_psychology.yaml
- mmlu_public_relations.yaml
- mmlu_security_studies.yaml
- mmlu_sociology.yaml
- mmlu_us_foreign_policy.yaml
- mmlu_virology.yaml
- mmlu_world_religions.yaml
- utils.py
- _mmlu.yaml
- _mmlu_flan_generative_template_yaml
- mmlu_abstract_algebra.yaml
- mmlu_anatomy.yaml
- mmlu_astronomy.yaml
- mmlu_business_ethics.yaml
- mmlu_clinical_knowledge.yaml
- mmlu_college_biology.yaml
- mmlu_college_chemistry.yaml
- mmlu_college_computer_science.yaml
- mmlu_college_mathematics.yaml
- mmlu_college_medicine.yaml
- mmlu_college_physics.yaml
- mmlu_computer_security.yaml
- mmlu_conceptual_physics.yaml
- mmlu_econometrics.yaml
- mmlu_electrical_engineering.yaml
- mmlu_elementary_mathematics.yaml
- mmlu_formal_logic.yaml
- mmlu_global_facts.yaml
- mmlu_high_school_biology.yaml
- mmlu_high_school_chemistry.yaml
- mmlu_high_school_computer_science.yaml
- mmlu_high_school_european_history.yaml
- mmlu_high_school_geography.yaml
- mmlu_high_school_government_and_politics.yaml
- mmlu_high_school_macroeconomics.yaml
- mmlu_high_school_mathematics.yaml
- mmlu_high_school_microeconomics.yaml
- mmlu_high_school_physics.yaml
- mmlu_high_school_psychology.yaml
- mmlu_high_school_statistics.yaml
- mmlu_high_school_us_history.yaml
- mmlu_high_school_world_history.yaml
- mmlu_human_aging.yaml
- mmlu_human_sexuality.yaml
- mmlu_international_law.yaml
- mmlu_jurisprudence.yaml
- mmlu_logical_fallacies.yaml
- mmlu_machine_learning.yaml
- mmlu_management.yaml
- mmlu_marketing.yaml
- mmlu_medical_genetics.yaml
- mmlu_miscellaneous.yaml
- mmlu_moral_disputes.yaml
- mmlu_moral_scenarios.yaml
- mmlu_nutrition.yaml
- mmlu_philosophy.yaml
- mmlu_prehistory.yaml
- mmlu_professional_accounting.yaml
- mmlu_professional_law.yaml
- mmlu_professional_medicine.yaml
- mmlu_professional_psychology.yaml
- mmlu_public_relations.yaml
- mmlu_security_studies.yaml
- mmlu_sociology.yaml
- mmlu_us_foreign_policy.yaml
- mmlu_virology.yaml
- mmlu_world_religions.yaml
- utils.py
- _mmlu.yaml
- _mmlu_flan_loglikelihood_template_yaml
- mmlu_abstract_algebra.yaml
- mmlu_anatomy.yaml
- mmlu_astronomy.yaml
- mmlu_business_ethics.yaml
- mmlu_clinical_knowledge.yaml
- mmlu_college_biology.yaml
- mmlu_college_chemistry.yaml
- mmlu_college_computer_science.yaml
- mmlu_college_mathematics.yaml
- mmlu_college_medicine.yaml
- mmlu_college_physics.yaml
- mmlu_computer_security.yaml
- mmlu_conceptual_physics.yaml
- mmlu_econometrics.yaml
- mmlu_electrical_engineering.yaml
- mmlu_elementary_mathematics.yaml
- mmlu_formal_logic.yaml
- mmlu_global_facts.yaml
- mmlu_high_school_biology.yaml
- mmlu_high_school_chemistry.yaml
- mmlu_high_school_computer_science.yaml
- mmlu_high_school_european_history.yaml
- mmlu_high_school_geography.yaml
- mmlu_high_school_government_and_politics.yaml
- mmlu_high_school_macroeconomics.yaml
- mmlu_high_school_mathematics.yaml
- mmlu_high_school_microeconomics.yaml
- mmlu_high_school_physics.yaml
- mmlu_high_school_psychology.yaml
- mmlu_high_school_statistics.yaml
- mmlu_high_school_us_history.yaml
- mmlu_high_school_world_history.yaml
- mmlu_human_aging.yaml
- mmlu_human_sexuality.yaml
- mmlu_international_law.yaml
- mmlu_jurisprudence.yaml
- mmlu_logical_fallacies.yaml
- mmlu_machine_learning.yaml
- mmlu_management.yaml
- mmlu_marketing.yaml
- mmlu_medical_genetics.yaml
- mmlu_miscellaneous.yaml
- mmlu_moral_disputes.yaml
- mmlu_moral_scenarios.yaml
- mmlu_nutrition.yaml
- mmlu_philosophy.yaml
- mmlu_prehistory.yaml
- mmlu_professional_accounting.yaml
- mmlu_professional_law.yaml
- mmlu_professional_medicine.yaml
- mmlu_professional_psychology.yaml
- mmlu_public_relations.yaml
- mmlu_security_studies.yaml
- mmlu_sociology.yaml
- mmlu_us_foreign_policy.yaml
- mmlu_virology.yaml
- mmlu_world_religions.yaml
- _default_template_yaml
- _mmlu.yaml
- mmlu_abstract_algebra.yaml
- mmlu_anatomy.yaml
- mmlu_astronomy.yaml
- mmlu_business_ethics.yaml
- mmlu_clinical_knowledge.yaml
- mmlu_college_biology.yaml
- mmlu_college_chemistry.yaml
- mmlu_college_computer_science.yaml
- mmlu_college_mathematics.yaml
- mmlu_college_medicine.yaml
- mmlu_college_physics.yaml
- mmlu_computer_security.yaml
- mmlu_conceptual_physics.yaml
- mmlu_econometrics.yaml
- mmlu_electrical_engineering.yaml
- mmlu_elementary_mathematics.yaml
- mmlu_formal_logic.yaml
- mmlu_global_facts.yaml
- mmlu_high_school_biology.yaml
- mmlu_high_school_chemistry.yaml
- mmlu_high_school_computer_science.yaml
- mmlu_high_school_european_history.yaml
- mmlu_high_school_geography.yaml
- mmlu_high_school_government_and_politics.yaml
- mmlu_high_school_macroeconomics.yaml
- mmlu_high_school_mathematics.yaml
- mmlu_high_school_microeconomics.yaml
- mmlu_high_school_physics.yaml
- mmlu_high_school_psychology.yaml
- mmlu_high_school_statistics.yaml
- mmlu_high_school_us_history.yaml
- mmlu_high_school_world_history.yaml
- mmlu_human_aging.yaml
- mmlu_human_sexuality.yaml
- mmlu_international_law.yaml
- mmlu_jurisprudence.yaml
- mmlu_logical_fallacies.yaml
- mmlu_machine_learning.yaml
- mmlu_management.yaml
- mmlu_marketing.yaml
- mmlu_medical_genetics.yaml
- mmlu_miscellaneous.yaml
- mmlu_moral_disputes.yaml
- mmlu_moral_scenarios.yaml
- mmlu_nutrition.yaml
- mmlu_philosophy.yaml
- mmlu_prehistory.yaml
- mmlu_professional_accounting.yaml
- mmlu_professional_law.yaml
- mmlu_professional_medicine.yaml
- mmlu_professional_psychology.yaml
- mmlu_public_relations.yaml
- mmlu_security_studies.yaml
- mmlu_sociology.yaml
- mmlu_us_foreign_policy.yaml
- mmlu_virology.yaml
- mmlu_world_religions.yaml
- _generate_configs.py
- _default_template_yaml
- _mmlu_pro.yaml
- mmlu_pro_biology.yaml
- mmlu_pro_business.yaml
- mmlu_pro_chemistry.yaml
- mmlu_pro_computer_science.yaml
- mmlu_pro_economics.yaml
- mmlu_pro_engineering.yaml
- mmlu_pro_health.yaml
- mmlu_pro_history.yaml
- mmlu_pro_law.yaml
- mmlu_pro_math.yaml
- mmlu_pro_other.yaml
- mmlu_pro_philosophy.yaml
- mmlu_pro_physics.yaml
- mmlu_pro_psychology.yaml
- README.md
- utils.py
- mmmu_val_reasoning.yaml
- utils.py
- _default_template_yaml
- arial.ttf
- mmmu.yaml
- mmmu_group_img.yaml
- mmmu_group_img_test.yaml
- mmmu_group_img_val.yaml
- mmmu_test.yaml
- mmmu_val.yaml
- mmmu_val_pass64.yaml
- mmmu_val_reasoning.yaml
- utils.py
- utils_group_img.py
- _default_template_yaml
- mmmu_pro_cot.yaml
- mmmu_pro_standard_cot.yaml
- mmmu_pro_vision_cot.yaml
- utils.py
- _default_template_yaml
- mmmu_pro.yaml
- mmmu_pro_composite.yaml
- mmmu_pro_composite_cot.yaml
- mmmu_pro_cot.yaml
- mmmu_pro_standard.yaml
- mmmu_pro_standard_cot.yaml
- mmmu_pro_vision.yaml
- mmmu_pro_vision_cot.yaml
- utils.py
- mmrefine.yaml
- mmrefine_evals.py
- prompts.py
- utils.py
- prompt.py
- prompt_w_imagesearch.py
- __init__.py
- tokenizers.py
- utils.py
- retriever.py
- f1_score.py
- req_score.py
- result_summary.py
- image_utils.py
- lmms_eval_utils.py
- prompt_utils.py
- utils.py
- web_content_utils.py
- constants.py
- get_final_scores.py
- lmms_eval_utils.py
- mmsearch.yaml
- mmsearch_end2end.yaml
- mmsearch_rerank.yaml
- mmsearch_summarization.yaml
- msr_bench.yaml
- utils.py
- _default_template_yaml
- mmstar.yaml
- mmstar_ko.yaml
- mmstar_oc.yaml
- utils.py
- _default_template_yaml
- mmt.yaml
- mmt_mi.yaml
- mmt_mi_test.yaml
- mmt_mi_val.yaml
- mmt_test.yaml
- mmt_val.yaml
- utils.py
- _default_template_mmupd_yaml
- mmaad_base.yaml
- mmaad_instruction.yaml
- mmaad_option.yaml
- mmiasd_base.yaml
- mmiasd_instruction.yaml
- mmiasd_option.yaml
- mmivqd_base.yaml
- mmivqd_instruction.yaml
- mmivqd_option.yaml
- mmupd.yaml
- mmupd_base.yaml
- mmupd_evals.py
- mmupd_instruction.yaml
- mmupd_option.yaml
- utils.py
- mmvet.yaml
- utils.py
- mmvetv2.yaml
- mmvetv2_group_img.yaml
- utils.py
- mmvu_val.yaml
- mmvu_val_cot copy.yaml
- utils.py
- mmworld.yaml
- utils.py
- _default_template_yaml
- moviechat_breakpoint.yaml
- moviechat_global.yaml
- README.md
- utils.py
- muchomusic.yaml
- utils.py
- muirbench.yaml
- utils.py
- multidocvqa.yaml
- multidocvqa_test.yaml
- multidocvqa_val.yaml
- utils.py
- _default_template_yaml
- arabic_llava_in_the_wild.yaml
- bengali_llava_in_the_wild.yaml
- chinese_llava_in_the_wild.yaml
- french_llava_in_the_wild.yaml
- hindi_llava_in_the_wild.yaml
- japanese_llava_in_the_wild.yaml
- README.md
- rule.json
- russian_llava_in_the_wild.yaml
- spanish_llava_in_the_wild.yaml
- urdu_llava_in_the_wild.yaml
- utils.py
- multimodal_rewardbench.yaml
- utils.py
- _default_template_yaml
- mvbench.yaml
- mvbench_action_antonym.yaml
- mvbench_action_count.yaml
- mvbench_action_localization.yaml
- mvbench_action_prediction.yaml
- mvbench_action_sequence.yaml
- mvbench_character_order.yaml
- mvbench_counterfactual_inference.yaml
- mvbench_egocentric_navigation.yaml
- mvbench_episodic_reasoning.yaml
- mvbench_fine_grained_action.yaml
- mvbench_fine_grained_pose.yaml
- mvbench_moving_attribute.yaml
- mvbench_moving_count.yaml
- mvbench_moving_direction.yaml
- mvbench_object_existence.yaml
- mvbench_object_interaction.yaml
- mvbench_object_shuffle.yaml
- mvbench_scene_transition.yaml
- mvbench_state_change.yaml
- mvbench_unexpected_action.yaml
- utils.py
- naturalbench.yaml
- utils.py
- _default_template_yaml
- nextqa.yaml
- nextqa_mc_test.yaml
- nextqa_oe_test.yaml
- nextqa_oe_val.yaml
- utils.py
- _default_template_nocaps_yaml
- nocaps.yaml
- nocaps_test.yaml
- nocaps_val.yaml
- nocaps_val_lite.yaml
- utils.py
- ocrbench.yaml
- upload_ocrbench.py
- utils.py
- ocrbench_v2_reasoning.yaml
- utils.py
- __init__.py
- rrc_evaluation_funcs_1_1.py
- script.py
- __init__.py
- IoUscore_metric.py
- ocrbench_v2.yaml
- page_ocr_metric.py
- parallel.py
- spotting_metric.py
- TEDS_metric.py
- upload_ocrbench_v2.py
- utils.py
- vqa_metric.py
- _default_template_vqa_yaml
- _generate_config.py
- _ok_vqa.yaml
- ok_vqa_val2014.yaml
- ok_vqa_val2014_lite.yaml
- utils.py
- cn_utils.py
- en_utils.py
- olympiadbench.yaml
- olympiadbench_evals.py
- olympiadbench_OE_MM_maths_en_COMP.yaml
- olympiadbench_OE_MM_physics_en_COMP.yaml
- testmini_utils.py
- olympiadbench_math_en_reasoning.yaml
- olympiadbench_math_zh_reasoning.yaml
- olympiadbench_physics_en_reasoning.yaml
- olympiadbench_physics_zh_reasoning.yaml
- olympiadbench_reasoning.yaml
- utils.py
- en_utils.py
- olympiadbench_all_boxed.yaml
- olympiadbench_boxed.yaml
- olympiadbench_evals.py
- olympiadbench_image_math_en.yaml
- olympiadbench_image_math_zh.yaml
- olympiadbench_image_physics_en.yaml
- olympiadbench_image_physics_zh.yaml
- olympiadbench_math_en.yaml
- olympiadbench_math_en_no_proof.yaml
- olympiadbench_math_zh.yaml
- olympiadbench_math_zh_no_proof.yaml
- olympiadbench_official.yaml
- olympiadbench_official_en.yaml
- olympiadbench_official_en_no_proof.yaml
- olympiadbench_official_no_proof.yaml
- olympiadbench_official_zh.yaml
- olympiadbench_official_zh_no_proof.yaml
- olympiadbench_physics_en.yaml
- olympiadbench_physics_en_no_proof.yaml
- olympiadbench_physics_zh.yaml
- olympiadbench_physics_zh_no_proof.yaml
- utils.py
- zh_utils.py
- _default_template_yaml
- omni_bench.yaml
- omni_bench_audio_transcript.yaml
- omni_bench_image_caption.yaml
- utils.py
- _default_template_yaml
- omnispatial.yaml
- utils.py
- _default_template_yaml
- openasr.yaml
- openasr_ami.yaml
- openasr_common_voice.yaml
- openasr_earnings22.yaml
- openasr_gigaspeech.yaml
- openasr_librispeech.yaml
- openasr_librispeech_test_clean.yaml
- openasr_librispeech_test_other.yaml
- openasr_spgispeech.yaml
- openasr_tedlium.yaml
- openasr_voxpopuli.yaml
- utils.py
- openai_math.yaml
- openai_math_agg64.yaml
- openai_math_cov64.yaml
- openai_math_cov64_train.yaml
- openai_math_maj64_cov64.yaml
- openai_math_maj64_cov64_train.yaml
- openai_math_train.yaml
- utils.py
- openhermes.yaml
- utils.py
- openslr_librispeech.yaml
- openslr_librispeech_other.yaml
- data.json
- data.json
- data.json
- README.md
- score.py
- constant.py
- ovo_backward.yaml
- ovo_forward.yaml
- ovo_realtime.yaml
- ovobench.yaml
- README.md
- utils.py
- people_speech_val.yaml
- utils.py
- _default_template_yaml
- perceptiontest_mc.yaml
- perceptiontest_mcppl.yaml
- utils.py
- _default_template_yaml
- perceptiontest_mc.yaml
- perceptiontest_mcppl.yaml
- utils.py
- _default_template_phyx_reasoning_yaml
- phyx_mc_reasoning.yaml
- phyx_mini_mc_reasoning.yaml
- phyx_mini_oe_reasoning.yaml
- phyx_oe_reasoning.yaml
- phyx_reasoning.yaml
- utils.py
- phyx.yaml
- phyx_evals.py
- phyx_mc.yaml
- phyx_mini_mc.yaml
- phyx_mini_oe.yaml
- phyx_oe.yaml
- utils.py
- fgqa_test.yaml
- fgqa_utils.py
- rcap_test.yaml
- rcap_utils.py
- rdcap_test.yaml
- rdcap_utils.py
- rtloc_test.yaml
- rtloc_utils.py
- sgqa_test.yaml
- sgqa_utils.py
- _default_template_yaml
- eval_utils.py
- README.md
- pope.yaml
- pope_adv.yaml
- pope_full.yaml
- pope_pop.yaml
- pope_random.yaml
- utils.py
- abench_dev.yaml
- qbench2_dev.yaml
- qbench_dev.yaml
- qbenchs_dev.yaml
- utils.py
- realworldqa_reasoning.yaml
- utils.py
- realworldqa.yaml
- utils.py
- _default_template_bbox_rec_yaml
- _default_template_bbox_yaml
- _default_template_seg_yaml
- _generate_config.py
- _refcoco.yaml
- refcoco_bbox_rec_test.yaml
- refcoco_bbox_rec_testA.yaml
- refcoco_bbox_rec_testB.yaml
- refcoco_bbox_rec_val.yaml
- refcoco_bbox_test.yaml
- refcoco_bbox_testA.yaml
- refcoco_bbox_testB.yaml
- refcoco_bbox_val.yaml
- refcoco_bbox_val_lite.yaml
- refcoco_seg_test.yaml
- refcoco_seg_testA.yaml
- refcoco_seg_testB.yaml
- refcoco_seg_val.yaml
- utils.py
- utils_rec.py
- _default_template_bbox_rec_yaml
- _default_template_bbox_yaml
- _default_template_seg_yaml
- _generate_config.py
- _refcoco.yaml
- refcoco+_bbox_rec_testA.yaml
- refcoco+_bbox_rec_testB.yaml
- refcoco+_bbox_rec_val.yaml
- refcoco+_bbox_testA.yaml
- refcoco+_bbox_testB.yaml
- refcoco+_bbox_val.yaml
- refcoco+_seg_testA.yaml
- refcoco+_seg_testB.yaml
- refcoco+_seg_val.yaml
- utils.py
- utils_rec.py
- _default_template_bbox_rec_yaml
- _default_template_bbox_yaml
- _default_template_seg_yaml
- _generate_config.py
- _refcoco.yaml
- refcocog_bbox_rec_test.yaml
- refcocog_bbox_rec_val.yaml
- refcocog_bbox_test.yaml
- refcocog_bbox_val.yaml
- refcocog_seg_test.yaml
- refcocog_seg_val.yaml
- utils.py
- utils_rec.py
- _default_template_yaml
- refspatial.yaml
- refspatial_location.yaml
- refspatial_placement.yaml
- refspatial_unseen.yaml
- utils.py
- _o3_default
- _p3_default
- o3.yaml
- o3_box.yaml
- o3_box_img.yaml
- p3.yaml
- p3_box.yaml
- p3_box_img.yaml
- utils.py
- scibench.yaml
- scibench_chemistry.yaml
- scibench_math.yaml
- scibench_physics.yaml
- utils.py
- scienceqa.yaml
- scienceqa_full.yaml
- scienceqa_img.yaml
- utils.py
- scivideobench.yaml
- utils.py
- _default_template_rec_yaml
- _default_template_reg_yaml
- _screenspot.yaml
- README.md
- screenspot_rec_test.yaml
- screenspot_reg_test.yaml
- utils.py
- utils_rec.py
- ko_utils.py
- seedbench.yaml
- seedbench_ko.yaml
- seedbench_lite.yaml
- seedbench_ppl.yaml
- utils.py
- seedbench_2.yaml
- utils.py
- seedbench_2_plus.yaml
- utils.py
- seephys.yaml
- seephys_evals.py
- seephys_utils.py
- site_video_multiimage.yaml
- utils.py
- site_image.yaml
- site_video.yaml
- utils.py
- metrics.py
- snsbench.yaml
- utils.py
- _default_template_yaml
- sparbench.yaml
- sparbench_tiny.yaml
- utils.py
- __init__.py
- cogmap_evaluator.py
- cogmap_metrics.py
- graph_operations.py
- __init__.py
- base_metrics.py
- extractors.py
- io_utils.py
- __init__.py
- evaluator.py
- __init__.py
- io_utils.py
- spatial_utils.py
- text_utils.py
- __init__.py
- __init__.py
- parse_output.py
- result_init.py
- rule_metrics.py
- spatialtreebench.yaml
- spatree_hierarchy.json
- utils.py
- _default_template_yaml
- spatialviz.yaml
- utils.py
- _default_template_yaml
- stare.yaml
- stare_2d_text_instruct.yaml
- stare_2d_text_instruct_vsim.yaml
- stare_2d_va.yaml
- stare_2d_va_vsim.yaml
- stare_3d_text_instruct.yaml
- stare_3d_text_instruct_vsim.yaml
- stare_3d_va.yaml
- stare_3d_va_vsim.yaml
- stare_folding_nets.yaml
- stare_folding_nets_vsim.yaml
- stare_perspective.yaml
- stare_tangram_puzzle.yaml
- stare_tangram_puzzle_vsim.yaml
- stare_temporal.yaml
- utils.py
- _default_template_yaml
- step2_audio_paralinguistic.yaml
- step2_audio_paralinguistic_age.yaml
- step2_audio_paralinguistic_emotions.yaml
- step2_audio_paralinguistic_event.yaml
- step2_audio_paralinguistic_gender.yaml
- step2_audio_paralinguistic_pitch.yaml
- step2_audio_paralinguistic_rhythm.yaml
- step2_audio_paralinguistic_scene.yaml
- step2_audio_paralinguistic_speed.yaml
- step2_audio_paralinguistic_vocalsound.yaml
- step2_audio_paralinguistic_voice_styles.yaml
- step2_audio_paralinguistic_voice_tone.yaml
- utils.py
- __init__.py
- README.md
- structeditbench.yaml
- utils.py
- stvqa.yaml
- utils.py
- super_gpqa.yaml
- super_gpqa_multishot.yaml
- utils.py
- donut_evaluator.py
- synthdog.yaml
- synthdog_en.yaml
- synthdog_zh.yaml
- utils.py
- tedlium_dev_test.yaml
- tedlium_long_form.yaml
- utils.py
- _default_template_yaml
- _tempcompass.yaml
- tempcompass_caption_matching.yaml
- tempcompass_captioning.yaml
- tempcompass_mc.yaml
- tempcompass_yes_no.yaml
- utils.py
- temporalbench.yaml
- temporalbench_long_qa.yaml
- temporalbench_short_caption.yaml
- temporalbench_short_qa.yaml
- utils.py
- _default_template_textcaps_yaml
- textcaps.yaml
- textcaps_test.yaml
- textcaps_train.yaml
- textcaps_val.yaml
- textcaps_val_lite.yaml
- utils.py
- _default_template_textvqa_yaml
- _textvqa.yaml
- textvqa_test.yaml
- textvqa_val.yaml
- textvqa_val_lite.yaml
- utils.py
- timescope.yaml
- utils.py
- tomato.yaml
- utils.py
- ueval.yaml
- utils.py
- _vatex.yaml
- utils.py
- vatex_test.yaml
- vatex_val_zh.yaml
- _default_template_vcr_yaml
- utils.py
- vcr_wiki_en_easy.yaml
- vcr_wiki_en_easy_100.yaml
- vcr_wiki_en_easy_500.yaml
- vcr_wiki_en_hard.yaml
- vcr_wiki_en_hard_100.yaml
- vcr_wiki_en_hard_500.yaml
- vcr_wiki_zh_easy.yaml
- vcr_wiki_zh_easy_100.yaml
- vcr_wiki_zh_easy_500.yaml
- vcr_wiki_zh_hard.yaml
- vcr_wiki_zh_hard_100.yaml
- vcr_wiki_zh_hard_500.yaml
- _default_template_yaml
- background_test.yaml
- camera_test.yaml
- detailed_test.yaml
- main_object_test.yaml
- README.md
- short_test.yaml
- utils.py
- utils.py
- vibe_eval.yaml
- _default_template.yaml
- gpt_utils.py
- utils.py
- videott_all.yaml
- videott_all_audio.yaml
- videott_correct_leading_oe.yaml
- videott_no_leading_oe.yaml
- videott_paraphrase_oe.yaml
- videott_single_mc.yaml
- videott_single_mc_description.yaml
- videott_wrong_leading_oe.yaml
- _default_template_yaml
- README.md
- utils.py
- video_detail_description.yaml
- _default_template_yaml
- _videochatgpt.yaml
- utils.py
- videochatgpt_consistency.yaml
- videochatgpt_generic.yaml
- videochatgpt_temporal.yaml
- utils.py
- videoevalpro.yaml
- cot_postprocess.py
- cot_step_evaluation.py
- README.md
- utils.py
- videomathqa_mbin.yaml
- videomathqa_mbin_cot.yaml
- videomathqa_mbin_cot_w_subtitle.yaml
- videomathqa_mbin_w_subtitle.yaml
- videomathqa_mcq.yaml
- videomathqa_mcq_cot.yaml
- videomathqa_mcq_cot_w_subtitle.yaml
- videomathqa_mcq_w_subtitle.yaml
- utils.py
- videomme_convert_mcq_oe.yaml
- utils.py
- videomme_gt_none_option.yaml
- utils.py
- videomme_no_visual.yaml
- utils.py
- videomme_number_option.yaml
- utils.py
- videomme_random_choice.yaml
- utils.py
- videomme_revert_oe_mcq.yaml
- utils.py
- videomme_video_only_abcd.yaml
- _default_template_yaml
- utils.py
- videomme.yaml
- videomme_long.yaml
- videomme_long_w_subtitle.yaml
- videomme_w_subtitle.yaml
- _default_template_yaml
- adaptation.yaml
- comprehension.yaml
- perception.yaml
- utils.py
- _default_template_yaml
- adaptation.yaml
- comprehension.yaml
- perception.yaml
- utils.py
- _default_template_yaml
- adaptation.yaml
- comprehension.yaml
- perception.yaml
- utils.py
- _default_template_yaml
- adaptation.yaml
- comprehension.yaml
- perception.yaml
- utils.py
- _default_template_yaml
- adaptation.yaml
- adaptation_question_only.yaml
- comprehension.yaml
- perception.yaml
- utils.py
- video_mmmu.yaml
- utils.py
- viewspatial.yaml
- utils.py
- vinoground.yaml
- utils.py
- VisualPuzzles_cot.yaml
- VisualPuzzles_direct.yaml
- prompts.py
- utils.py
- visualwebbench_action_ground.yaml
- visualwebbench_action_prediction.yaml
- visualwebbench_element_ground.yaml
- visualwebbench_element_ocr.yaml
- visualwebbench_heading_ocr.yaml
- visualwebbench_web_caption.yaml
- visualwebbench_webqa.yaml
- _default_template_yaml
- _vitatecs.yaml
- utils.py
- vitatecs_compositionality.yaml
- vitatecs_direction.yaml
- vitatecs_intensity.yaml
- vitatecs_localization.yaml
- vitatecs_sequence.yaml
- vitatecs_type.yaml
- _default_template_vqa_yaml
- _generate_config.py
- _vizwiz_vqa.yaml
- utils.py
- vizwiz_vqa_test.yaml
- vizwiz_vqa_val.yaml
- vizwiz_vqa_val_lite.yaml
- utils.py
- vl_rewardbench.yaml
- __init__.py
- README.md
- utils.py
- vlms_are_biased.yaml
- __init__.py
- README.md
- utils.py
- vlmsareblind.yaml
- vlmsareblind_lite.yaml
- utils.py
- vmcbench.yaml
- _default_template_yaml
- utils.py
- vocalsound_test.yaml
- vocalsound_val.yaml
- __init__.py
- instructions.py
- instructions_registry.py
- instructions_util.py
- _default_template_yaml
- utils.py
- voicebench.yaml
- voicebench_advbench.yaml
- voicebench_alpacaeval.yaml
- voicebench_bbh.yaml
- voicebench_commoneval.yaml
- voicebench_ifeval.yaml
- voicebench_mmsu.yaml
- voicebench_mmsu_biology.yaml
- voicebench_mmsu_business.yaml
- voicebench_mmsu_chemistry.yaml
- voicebench_mmsu_economics.yaml
- voicebench_mmsu_engineering.yaml
- voicebench_mmsu_health.yaml
- voicebench_mmsu_history.yaml
- voicebench_mmsu_law.yaml
- voicebench_mmsu_other.yaml
- voicebench_mmsu_philosophy.yaml
- voicebench_mmsu_physics.yaml
- voicebench_mmsu_psychology.yaml
- voicebench_openbookqa.yaml
- voicebench_sd-qa.yaml
- voicebench_sd-qa_aus.yaml
- voicebench_sd-qa_gbr.yaml
- voicebench_sd-qa_ind_n.yaml
- voicebench_sd-qa_ind_s.yaml
- voicebench_sd-qa_irl.yaml
- voicebench_sd-qa_kenya.yaml
- voicebench_sd-qa_nga.yaml
- voicebench_sd-qa_nzl.yaml
- voicebench_sd-qa_phl.yaml
- voicebench_sd-qa_usa.yaml
- voicebench_sd-qa_zaf.yaml
- voicebench_wildvoice.yaml
- _default_template_vqav2_yaml
- _vqav2.yaml
- utils.py
- vqav2_test.yaml
- vqav2_val.yaml
- vqav2_val_lite.yaml
- utils.py
- vsibench_multiimage.yaml
- _default_template_yaml
- utils.py
- vsibench.yaml
- vsibench_debiased.yaml
- vsibench_pruned.yaml
- __init__.py
- _default_template_yaml
- README.md
- utils.py
- vstar_bench.yaml
- vstar_bench_direct_attributes.yaml
- vstar_bench_relative_position.yaml
- utils.py
- wavcaps.yaml
- README.md
- utils.py
- websrc.yaml
- websrc_test.yaml
- websrc_val.yaml
- utils.py
- wemath_testmini_thinking.yaml
- wemath_utils.py
- _default_template_yaml
- utils.py
- wenet_speech.yaml
- wenet_speech_dev.yaml
- wenet_speech_test_meeting.yaml
- _default_template_yaml
- utils.py
- where2place.yaml
- _default_template_yaml
- utils.py
- wild_vision_bench0617.yaml
- wild_vision_bench0630.yaml
- wildvision_bench.yaml
- _default_template_yaml
- utils.py
- worldqa.yaml
- worldqa_generation.yaml
- worldqa_mc.yaml
- worldqa_mc_evaluator.py
- worldqa_mcppl.yaml
- utils.py
- worldsense.yaml
- worldsense_w_subtitle.yaml
- mcq_utils.py
- XLRS-lite.yaml
- _default_template_yaml
- utils.py
- youcook2_val.yaml
- __init__.py
- index.html
- App.tsx
- index.css
- main.tsx
- index.html
- package-lock.json
- package.json
- postcss.config.js
- tailwind.config.js
- tsconfig.json
- vite.config.ts
- __init__.py
- cli.py
- discovery.py
- README.md
- server.py
- __init__.py
- __main__.py
- evaluator.py
- evaluator_utils.py
- imports.py
- launch_server.py
- logging_utils.py
- protocol.py
- utils.py
- example_eval.yaml
- llava_result_check.md
- llava_sglang_result_check.md
- script.sh
- test_llava.py
- test_scienceqa.py
- __init__.py
- test_qwen2_5_vl.py
- __init__.py
- run_cicd.py
- utils.py
- get_split_zip.py
- make_image_hf_dataset.ipynb
- README.md
- regression.py
- .gitignore
- .pre-commit-config.yaml
- CLAUDE.md
- LICENSE
- pyproject.toml
- README.md
- setup.py
- eval_ov_encoder.sh
- preprocess_video_benchmark.sh
- run_offline_codec_patch_for_eval.sh
- run_stage1.sh
- run_stage2.sh
- pretrain_ov_encoder.sh
- pretrain_siglip2.sh
- sft_ov_encoder.sh
- sft_siglip2.sh
- zero2_noopt.json
- zero3.json
- .dockerignore
- dockerfile
- pyproject.toml
- README.md
- __init__.py
- layers.py
- vit_aim_v2.py
- vit_clip.py
- vit_dinov2.py
- vit_dinov3.py
- vit_metaclip.py
- vit_ov_encoder.py
- vit_siglip.py
- vit_siglip2.py
- __init__.py
- configuration_onevision_encoder.py
- modeling_onevision_encoder.py
- README.md
- README_lang.md
- ov_encoder_base_stage1_si.sh
- ov_encoder_base_stage1_si_dry_run.sh
- ov_encoder_huge_stage2_residual_128gpus.sh
- ov_encoder_large_stage2_residual_128gpus.sh
- ov_encoder_large_stage2_residual_8gpus.sh
- __init__.py
- conftest.py
- test_automodel_consistency.py
- test_consistency.py
- test_onevision_encoder.py
- cat_npy_to_one.py
- check_video_parallel.py
- run_step1.sh
- run_step3.sh
- run_step4.sh
- split_recs.py
- step1_extract_video_features.py
- step2_compact_features.py
- step3_kmeans.py
- step4_collision.py
- video_list_to_tar.py
- __init__.py
- hevc_feature_decoder.py
- hevc_feature_decoder_mv.py
- step1_offline_preprocess.py
- step2_convert_videos2hevc.py
- step3_generate_video_mv_residual_index.py
- step3_generate_video_residual_index.py
- cp_to_ssd_jsons.py
- cp_to_ssd_obelics.py
- cp_to_ssd_zero250m.py
- pack_ocr.py
- split_recs.py
- inverted_index.py
- main_streamlit.py
- split_np.py
- __init__.py
- checkpoint_utils.py
- fused_partial_fc_v2.py
- fused_partial_fc_v2_multi_res.py
- lr_scheduler.py
- train.py
- .gitignore
- .pylintrc
- dockerfile
- environment.yml
- LICENSE
- pyproject.toml
- README.md
- requirements.txt
# Installation Guide
git clone https://github.com/EvolvingLMMs-Lab/OneVision-Encoder
Downloads the entire project code from GitHub to your computer.
cd OneVision-Encoder
Moves into the project folder you just downloaded.
2. Official Install Script
Easy Recommended- Python 3 Python is required to use pip.
pip install torch==2.7.0 torchvision==0.22.0 torchaudio==2.7.0 --index-url https://download.pytorch.org/whl/cu118
Installs the package published on PyPI directly โ no need to clone the source.
pip install --extra-index-url https://pypi.nvidia.com --upgrade nvidia-dali-cuda110
Installs the package published on PyPI directly โ no need to clone the source.
Pulled directly from this repo's README.
3. Docker
Easy- Git Needed to download the project code from GitHub.
- Docker Desktop Needed to build and run containers. Install it and keep it running in the background.
docker build -t ov_encoder_llava:26.01 .
Builds a runnable image based on the Dockerfile.
docker run -it --gpus all --ipc host --net host --privileged \
Runs the built image as an actual container.
docker build -t onevision-encoder:2601 .
Builds a runnable image based on the Dockerfile.
docker run -it --rm --gpus all --ipc host --net host --privileged \
Runs the built image as an actual container.
Pulled directly from this repo's README.
4. Node.js
Easycd llava_next/lmms-eval/lmms_eval/tui/web
This project's files live in a subfolder, so move into it first.
npm install
Downloads and installs the libraries listed in package.json.
npm start
Starts the development/run server.
5. Python
Easypip install -e .
Installs the Python libraries listed in requirements.txt (or similar).
pip install torch==2.7.0 torchvision==0.22.0 torchaudio==2.7.0 --index-url https://download.pytorch.org/whl/cu118
Installs the package published on PyPI directly โ no need to clone the source.
pip install --extra-index-url https://pypi.nvidia.com --upgrade nvidia-dali-cuda110
Installs the package published on PyPI directly โ no need to clone the source.
pip install -r requirements.txt
Installs the Python libraries listed in requirements.txt (or similar).
Pulled directly from this repo's README.
