transformers
🤗 Transformers: 텍스트, 비전, 오디오, 멀티모달 분야의 최첨단(state-of-the-art) 머신러닝 모델을 추론(inference)과 학습(training) 모두에서 다룰 수 있도록 지원하는 모델 정의 프레임워크입니다.
파일 탐색기
최종 버전 다운로드 (.zip)파일 수가 많아 일부만 표시됩니다. 전체 파일은 위 다운로드 버튼으로 확인해 주세요.
- SKILL.md
- AGENTS.md
- review-rules.md
- config.yml
- config.yml.bak
- create_circleci_config.py
- parse_test_outputs.py
- TROUBLESHOOT.md
- build.sh
- meta.yaml
- bug-report.yml
- config.yml
- feature-request.yml
- i18n.md
- migration.yml
- new-model-addition.yml
- codeowners_for_review_action
- add-model-like.yml
- ai-review.yml
- anti-slop.yml
- assign-reviewers.yml
- benchmark.yml
- benchmark_v2.yml
- benchmark_v2_a10_caller.yml
- benchmark_v2_mi300_caller.yml
- build-ci-docker-images.yml
- build-docker-images.yml
- build-nightly-ci-docker-images.yml
- build-past-ci-docker-images.yml
- build_documentation.yml
- build_pr_documentation.yml
- check-workflow-permissions.yml
- check_failed_tests.yml
- check_tiny_models.yml
- circleci-failure-summary-comment.yml.bak
- codeql.yml
- collated-reports.yml
- doctest_job.yml
- doctests.yml
- extras-smoke-test.yml
- get-pr-info.yml
- get-pr-number.yml
- model_jobs_intel_gaudi.yml
- new_model_pr_merged_notification.yml
- nightly-integration-failure-triage-caller.yml
- pr-ci-caller.yml
- pr-ci-post-dashboard-link.yml
- pr-ci-security-gate.yml
- pr-repo-consistency-bot.yml
- pr_build_doc_with_comment.yml
- pr_slow_ci_suggestion.yml
- push-important-models.yml
- release-conda.yml
- release.yml
- self-comment-ci.yml
- self-nightly-caller.yml
- self-nightly-past-ci-caller.yml
- self-past-caller.yml
- self-scheduled-amd-caller.yml
- self-scheduled-amd-mi250-caller.yml
- self-scheduled-amd-mi300-caller.yml
- self-scheduled-amd-mi355-caller.yml
- self-scheduled-caller.yml
- self-scheduled-flash-attn-caller.yml
- self-scheduled-intel-gaudi.yml
- self-scheduled-intel-gaudi3-caller.yml
- serge-verify-caller.yml
- slack-report.yml
- ssh-runner.yml
- stale.yml
- trl-ci-bot.yml
- TROUBLESHOOT.md
- trufflehog.yml
- update_metdata.yml
- upload_pr_documentation.yml
- vllm-ci-caller.yml
- copilot-instructions.md
- dependabot.yml
- PULL_REQUEST_TEMPLATE.md
- llama.py
- generation.yaml
- init_db.sql
- .gitignore
- __init__.py
- benchmark.py
- benchmarks_entrypoint.py
- default.yml
- grafana_dashboard.json
- grafana_datasource.yaml
- optimum_benchmark_wrapper.py
- README.md
- requirements.txt
- continuous_batching_overall.py
- benchmark_config.py
- benchmark_runner.py
- data_classes.py
- hardware_metrics.py
- .gitignore
- README.md
- requirements.txt
- run_benchmarks.py
- Dockerfile
- Dockerfile
- Dockerfile
- Dockerfile
- Dockerfile
- Dockerfile
- Dockerfile
- Dockerfile
- Dockerfile
- bert-base-cased.jsonnet
- dataset.yaml
- docker-entrypoint.sh
- Dockerfile
- Dockerfile
- Dockerfile
- consistency.dockerfile
- custom-tokenizers.dockerfile
- examples-torch.dockerfile
- exotic-models.dockerfile
- pipeline-torch.dockerfile
- quality.dockerfile
- README.md
- torch-light.dockerfile
- language_modeling.md
- masked_language_modeling.md
- multiple_choice.md
- question_answering.md
- sequence_classification.md
- summarization.md
- token_classification.md
- translation.md
- _config.py
- _toctree.yml
- accelerate.md
- attention.md
- autoclass_tutorial.md
- bertology.md
- chat_templating.md
- community.md
- conversations.md
- create_a_model.md
- custom_models.md
- fast_tokenizers.md
- gguf.md
- glossary.md
- how_to_hack_models.md
- index.md
- installation.md
- llm_tutorial.md
- llm_tutorial_optimization.md
- model_memory_anatomy.md
- model_sharing.md
- model_summary.md
- modular_transformers.md
- multilingual.md
- notebooks.md
- pad_truncation.md
- peft.md
- perplexity.md
- philosophy.md
- pipeline_tutorial.md
- pipeline_webserver.md
- preprocessing.md
- quicktour.md
- run_scripts.md
- sagemaker.md
- serialization.md
- task_summary.md
- tasks_explained.md
- tiktoken.md
- tokenizer_summary.md
- trainer.md
- training.md
- troubleshooting.md
- _config.py
- _toctree.yml
- accelerate.md
- add_new_model.md
- add_new_pipeline.md
- autoclass_tutorial.md
- contributing.md
- index.md
- installation.md
- llm_tutorial.md
- model_sharing.md
- peft.md
- pipeline_tutorial.md
- pr_checks.md
- preprocessing.md
- quicktour.md
- run_scripts.md
- testing.md
- training.md
- axolotl.md
- candle.md
- executorch.md
- llama_cpp.md
- mlx.md
- nanotron.md
- nemo_automodel_finetuning.md
- nemo_automodel_pretraining.md
- sglang.md
- tensorrt-llm.md
- torchtitan.md
- transformers_as_backend.md
- trl.md
- unsloth.md
- vllm.md
- audio_utils.md
- file_utils.md
- generation_utils.md
- image_processing_utils.md
- import_utils.md
- model_debugging_utils.md
- modeling_utils.md
- pipelines_utils.md
- rope_utils.md
- time_series_utils.md
- tokenization_utils.md
- trainer_utils.md
- loading_kernels.md
- overview.md
- writing_kernels.md
- backbones.md
- callback.md
- configuration.md
- continuous_batching.md
- data_collator.md
- deepspeed.md
- executorch.md
- exporters.md
- feature_extractor.md
- image_processor.md
- kernels.md
- logging.md
- model.md
- optimizer_schedules.md
- output.md
- peft.md
- pipelines.md
- processors.md
- quantization.md
- text_generation.md
- tokenizer.md
- trainer.md
- video_processor.md
- afmoe.md
- aimv2.md
- albert.md
- align.md
- altclip.md
- apertus.md
- arcee.md
- aria.md
- audio-spectrogram-transformer.md
- audioflamingo3.md
- auto.md
- autoformer.md
- axk1.md
- axk2.md
- aya_vision.md
- bamba.md
- bark.md
- bart.md
- barthez.md
- bartpho.md
- beit.md
- bert-generation.md
- bert-japanese.md
- bert.md
- bertweet.md
- big_bird.md
- bigbird_pegasus.md
- biogpt.md
- bit.md
- bitnet.md
- blenderbot-small.md
- blenderbot.md
- blip-2.md
- blip.md
- bloom.md
- blt.md
- bridgetower.md
- bros.md
- byt5.md
- camembert.md
- canine.md
- chameleon.md
- chinese_clip.md
- chmv2.md
- clap.md
- clip.md
- clipseg.md
- clvp.md
- code_llama.md
- codegen.md
- cohere.md
- cohere2.md
- cohere2_moe.md
- cohere2_vision.md
- cohere_asr.md
- cohere_compass.md
- colmodernvbert.md
- colpali.md
- colqwen2.md
- conditional_detr.md
- convbert.md
- convnext.md
- convnextv2.md
- cosmos3_edge.md
- cosmos3_omni.md
- cpm.md
- cpmant.md
- csm.md
- ctrl.md
- cvt.md
- cwm.md
- d_fine.md
- dab-detr.md
- dac.md
- data2vec.md
- dbrx.md
- deberta-v2.md
- deberta.md
- decision_transformer.md
- deepseek_ocr2.md
- deepseek_v2.md
- deepseek_v3.md
- deepseek_v32.md
- deepseek_v4.md
- deepseek_vl.md
- deepseek_vl_hybrid.md
- deformable_detr.md
- deimv2.md
- deit.md
- deplot.md
- depth_anything.md
- depth_anything_v2.md
- depth_pro.md
- detr.md
- dia.md
- diffllama.md
- diffusion_gemma.md
- dinat.md
- dinov2.md
- dinov2_with_registers.md
- dinov3.md
- distilbert.md
- doge.md
- donut.md
- dots1.md
- dpr.md
- dpt.md
- edgetam.md
- edgetam_video.md
- efficientloftr.md
- efficientnet.md
- electra.md
- emu3.md
- encodec.md
- encoder-decoder.md
- eomt.md
- eomt_dinov3.md
- ernie.md
- ernie4_5.md
- ernie4_5_moe.md
- ernie4_5_vl_moe.md
- esm.md
- esmc.md
- esmfold2.md
- eurobert.md
- evolla.md
- exaone4.md
- exaone4_5.md
- exaone_moe.md
- falcon.md
- falcon3.md
- falcon_h1.md
- falcon_mamba.md
- fast_vlm.md
- fastspeech2_conformer.md
- flan-t5.md
- flan-ul2.md
- flaubert.md
- flava.md
- flex_olmo.md
- florence2.md
- fnet.md
- focalnet.md
- fsmt.md
- funnel.md
- fuyu.md
- gemma.md
- gemma2.md
- gemma3.md
- gemma3n.md
- gemma4.md
- gemma4_assistant.md
- gemma4_unified.md
- gemma4_unified_assistant.md
- git.md
- glm.md
- glm4.md
- glm46v.md
- glm4_moe.md
- glm4_moe_lite.md
- glm4v.md
- glm4v_moe.md
- glm_image.md
- glm_moe_dsa.md
- glm_ocr.md
- glmasr.md
- glmga.md
- glpn.md
- got_ocr2.md
- gpt-sw3.md
- gpt2.md
- gpt_bigcode.md
- gpt_neo.md
- gpt_neox.md
- gpt_neox_japanese.md
- gpt_oss.md
- gptj.md
- granite.md
- granite4_vision.md
- granite_speech.md
- granite_speech_plus.md
- granite_swa.md
- granitemoe.md
- granitemoe_swa.md
- granitemoehybrid.md
- granitemoeshared.md
- granitevision.md
- grounding-dino.md
- groupvit.md
- helium.md
- herbert.md
- hgnet_v2.md
- hiera.md
- higgs_audio_v2.md
- higgs_audio_v2_tokenizer.md
- hrm_text.md
- hubert.md
- hunyuan_v1_dense.md
- hunyuan_v1_moe.md
- hunyuan_vl.md
- hy_v3.md
- hyperclovax.md
- ibert.md
- idefics.md
- idefics2.md
- idefics3.md
- ijepa.md
- imagegpt.md
- informer.md
- inkling.md
- instructblip.md
- instructblipvideo.md
- internvl.md
- jais2.md
- jamba.md
- janus.md
- jetmoe.md
- jina_embeddings_v3.md
- kimi_k25.md
- kosmos-2.md
- kosmos2_5.md
- kyutai_speech_to_text.md
- laguna.md
- lasr.md
- layoutlm.md
- layoutlmv2.md
- layoutlmv3.md
- layoutxlm.md
- led.md
- levit.md
- lfm2.md
- lfm2_moe.md
- lfm2_vl.md
- lightglue.md
- lighton_ocr.md
- lilt.md
- llama.md
- llama2.md
- llama3.md
- llama4.md
- llava.md
- llava_next.md
- llava_next_video.md
- llava_onevision.md
- longcat_flash.md
- longformer.md
- longt5.md
- luke.md
- lw_detr.md
- lxmert.md
- m2m_100.md
- madlad-400.md
- mamba.md
- mamba2.md
- marian.md
- markuplm.md
- mask2former.md
- maskformer.md
- matcha.md
- mbart.md
- megatron-bert.md
- mellum.md
- metaclip_2.md
- mgp-str.md
- mimi.md
- mimo_v2_flash.md
- minicpm3.md
- minicpmv4_6.md
- minimax.md
- minimax_m2.md
- minimax_m3_vl.md
- ministral.md
- ministral3.md
- mistral.md
- mistral3.md
- mistral4.md
- mixtral.md
- mlcd.md
- mllama.md
- mluke.md
- mm-grounding-dino.md
- mms.md
- mobilebert.md
- mobilenet_v1.md
- mobilenet_v2.md
- mobilevit.md
- mobilevitv2.md
- modernbert-decoder.md
- modernbert.md
- modernvbert.md
- moonshine.md
- moonshine_streaming.md
- moshi.md
- mpnet.md
- mpt.md
- mra.md
- mt5.md
- muse_glimmer.md
- muse_glimmer_assistant.md
- musicflamingo.md
- musicgen.md
- musicgen_melody.md
- mvp.md
- myt5.md
- nanochat.md
- nemotron.md
- nemotron3_5_asr.md
- nemotron_asr_streaming.md
- nemotron_h.md
- nllb-moe.md
- nllb.md
- nomic_bert.md
- nougat.md
- nystromformer.md
- olmo.md
- olmo2.md
- olmo3.md
- olmo_hybrid.md
- olmoe.md
- omdet-turbo.md
- oneformer.md
- openai-gpt.md
- openai_privacy_filter.md
- opt.md
- ovis2.md
- owlv2.md
- owlvit.md
- paddleocr_vl.md
- paligemma.md
- parakeet.md
- patchtsmixer.md
- patchtst.md
- pe_audio.md
- pe_audio_video.md
- pe_video.md
- pegasus.md
- pegasus_x.md
- perceiver.md
- perception_lm.md
- persimmon.md
- phi.md
- phi3.md
- phi4_multimodal.md
- phimoe.md
- phobert.md
- pi0.md
- pix2struct.md
- pixio.md
- pixtral.md
- plbart.md
- poolformer.md
- pop2piano.md
- pp_chart2table.md
- pp_doclayout_v2.md
- pp_doclayout_v3.md
- pp_formulanet.md
- pp_lcnet.md
- pp_lcnet_v3.md
- pp_lcnet_v4.md
- pp_ocrv5_mobile_det.md
- pp_ocrv5_mobile_rec.md
- pp_ocrv5_server_det.md
- pp_ocrv5_server_rec.md
- pp_ocrv6_medium_det.md
- pp_ocrv6_small_det.md
- pp_ocrv6_small_rec.md
- pp_ocrv6_tiny_rec.md
- prompt_depth_anything.md
- prophetnet.md
- pvt.md
- pvt_v2.md
- qianfan_ocr.md
- qwen2.md
- qwen2_5_omni.md
- qwen2_5_vl.md
- qwen2_audio.md
- qwen2_moe.md
- qwen2_vl.md
- qwen3.md
- qwen3_5.md
- qwen3_5_moe.md
- qwen3_asr.md
- qwen3_moe.md
- qwen3_next.md
- qwen3_omni_moe.md
- qwen3_vl.md
- qwen3_vl_moe.md
- radio.md
- rag.md
- recurrent_gemma.md
- reformer.md
- regnet.md
- rembert.md
- resnet.md
- rf_detr.md
- roberta-prelayernorm.md
- roberta.md
- roc_bert.md
- roformer.md
- rt_detr.md
- rt_detr_v2.md
- rwkv.md
- sam.md
- sam2.md
- sam2_video.md
- sam3.md
- sam3_lite_text.md
- sam3_tracker.md
- sam3_tracker_video.md
- sam3_video.md
- sam_hq.md
- sapiens2.md
- seamless_m4t.md
- seamless_m4t_v2.md
- seed_oss.md
- segformer.md
- seggpt.md
- sew-d.md
- sew.md
- shieldgemma2.md
- siglip.md
- siglip2.md
- slanet.md
- slanext.md
- smollm3.md
- smolvlm.md
- solar_open.md
- speech-encoder-decoder.md
- speech_to_text.md
- speecht5.md
- splinter.md
- squeezebert.md
- stablelm.md
- starcoder2.md
- step3p7.md
- superglue.md
- superpoint.md
- swiftformer.md
- swin.md
- swin2sr.md
- swinv2.md
- switch_transformers.md
- t5.md
- t5gemma.md
- t5gemma2.md
- t5v1.1.md
- table-transformer.md
- tapas.md
- textnet.md
- time_series_transformer.md
- timesfm.md
- timesfm2_5.md
- timesformer.md
- timm_wrapper.md
- tipsv2.md
- tipsv2_dpt.md
- trocr.md
- tvp.md
- udop.md
- ul2.md
- umt5.md
- unispeech-sat.md
- unispeech.md
- univnet.md
- upernet.md
- uvdoc.md
- vaultgemma.md
- vibevoice_acoustic_tokenizer.md
- vibevoice_asr.md
- video_llama_3.md
- video_llava.md
- videomae.md
- videomt.md
- videoprism.md
- vilt.md
- vipllava.md
- vision-encoder-decoder.md
- vision-text-dual-encoder.md
- visual_bert.md
- vit.md
- vit_mae.md
- vit_msn.md
- vitdet.md
- vitmatte.md
- vitpose.md
- vits.md
- vivit.md
- vjepa2.md
- voxtral.md
- voxtral_realtime.md
- wav2vec2-bert.md
- wav2vec2-conformer.md
- wav2vec2.md
- wav2vec2_phoneme.md
- wavlm.md
- whisper.md
- xclip.md
- xcodec.md
- xcodec2.md
- xglm.md
- xlm-roberta-xl.md
- xlm-roberta.md
- xlm-v.md
- xlm.md
- xlnet.md
- xls_r.md
- xlsr_wav2vec2.md
- xlstm.md
- xmod.md
- yolos.md
- yoso.md
- youtu.md
- zamba.md
- zamba2.md
- zaya.md
- zoedepth.md
- aqlm.md
- auto_round.md
- awq.md
- bitnet.md
- bitsandbytes.md
- compressed_tensors.md
- concept_guide.md
- contribute.md
- eetq.md
- fbgemm_fp8.md
- finegrained_fp8.md
- fouroversix.md
- fp_quant.md
- gptq.md
- higgs.md
- hqq.md
- metal.md
- mxfp4.md
- nvfp4.md
- optimum.md
- overview.md
- quanto.md
- quark.md
- selecting.md
- sinq.md
- spqr.md
- torchao.md
- vptq.md
- environment_variables.md
- cursor.md
- jan.md
- openweb_ui.md
- serving.md
- serving_optims.md
- tiny_agents.md
- any_to_any.md
- asr.md
- audio_classification.md
- audio_text_to_text.md
- document_question_answering.md
- image_captioning.md
- image_classification.md
- image_feature_extraction.md
- image_text_to_text.md
- instance_segmentation.md
- keypoint_detection.md
- keypoint_matching.md
- knowledge_distillation_for_image_classification.md
- language_modeling.md
- mask_generation.md
- masked_language_modeling.md
- monocular_depth_estimation.md
- multiple_choice.md
- object_detection.md
- prompting.md
- question_answering.md
- semantic_segmentation.md
- sequence_classification.md
- summarization.md
- text-to-speech.md
- token_classification.md
- training_vision_backbone.md
- translation.md
- video_classification.md
- video_text_to_text.md
- visual_document_retrieval.md
- visual_question_answering.md
- zero_shot_image_classification.md
- zero_shot_object_detection.md
- _config.py
- _redirects.yml
- _toctree.yml
- accelerate.md
- accelerator_selection.md
- add_audio_processing_components.md
- add_new_model.md
- add_new_pipeline.md
- add_vision_processing_components.md
- assisted_decoding.md
- attention_interface.md
- auto_docstring.md
- backbones.md
- cache_explanation.md
- chat_content_patterns.md
- chat_extras.md
- chat_response_parsing.md
- chat_templating.md
- chat_templating_multimodal.md
- chat_templating_writing.md
- community.md
- continuous_batching.md
- continuous_batching_architecture.md
- contributing.md
- conversations.md
- custom_models.md
- custom_tokenizers.md
- data_collators.md
- ddp.md
- debugging.md
- deepspeed.md
- deepspeed_alst.md
- expert_parallelism.md
- experts_interface.md
- exporters.md
- exporters_extend.md
- fast_tokenizers.md
- feature_extractors.md
- fsdp.md
- fusion_mapping.md
- generation_features.md
- generation_strategies.md
- gguf.md
- glossary.md
- grad_accumulation.md
- grad_checkpointing.md
- heterogeneous_configurations.md
- how_to_hack_models.md
- hpo_train.md
- image_processors.md
- index.md
- installation.md
- kernels.md
- kv_cache.md
- llm_tutorial.md
- llm_tutorial_optimization.md
- mixed_precision_training.md
- model_memory_anatomy.md
- model_output_tracing.md
- model_sharing.md
- modeling_rules.md
- models.md
- models_timeline.md
- modular_transformers.md
- monkey_patching.md
- multimodal_processing.md
- notebooks.md
- optimization_overview.md
- optimizers.md
- padding_free.md
- paged_attention.md
- peft.md
- perf_hardware.md
- perf_infer_gpu_multi.md
- perf_torch_compile.md
- perf_train_cpu.md
- perf_train_gaudi.md
- perf_train_gpu_many.md
- perf_train_special.md
- perplexity.md
- philosophy.md
- pipeline_gradio.md
- pipeline_tutorial.md
- pipeline_webserver.md
- pr_checks.md
- processors.md
- quicktour.md
- run_scripts.md
- serialization.md
- tensor_parallelism.md
- testing.md
- tokenizer_summary.md
- torch_compile.md
- trainer.md
- trainer_callbacks.md
- trainer_customize.md
- trainer_recipes.md
- training.md
- troubleshooting.md
- video_processors.md
- weightconverter.md
- asr.md
- audio_classification.md
- image_captioning.md
- image_classification.md
- language_modeling.md
- multiple_choice.md
- question_answering.md
- summarization.md
- _config.py
- _toctree.yml
- accelerate.md
- add_new_pipeline.md
- attention.md
- autoclass_tutorial.md
- bertology.md
- chat_templating.md
- community.md
- conversations.md
- create_a_model.md
- custom_models.md
- debugging.md
- fast_tokenizers.md
- glossary.md
- index.md
- installation.md
- model_memory_anatomy.md
- model_sharing.md
- multilingual.md
- pad_truncation.md
- performance.md
- perplexity.md
- philosophy.md
- pipeline_tutorial.md
- pipeline_webserver.md
- pr_checks.md
- preprocessing.md
- quicktour.md
- run_scripts.md
- sagemaker.md
- task_summary.md
- tasks_explained.md
- tokenizer_summary.md
- torch_compile.md
- trainer.md
- training.md
- _config.py
- _toctree.yml
- autoclass_tutorial.md
- in_translation.md
- index.md
- installation.md
- quicktour.md
- run_scripts_fr.md
- task_summary.md
- tasks_explained.md
- tutoriel_pipeline.md
- _toctree.yml
- accelerate.md
- pipeline_tutorial.md
- _config.py
- _toctree.yml
- accelerate.md
- add_new_model.md
- add_new_pipeline.md
- autoclass_tutorial.md
- big_models.md
- community.md
- create_a_model.md
- custom_models.md
- debugging.md
- index.md
- installation.md
- migration.md
- model_sharing.md
- multilingual.md
- perf_hardware.md
- perf_infer_cpu.md
- perf_infer_gpu_many.md
- perf_infer_gpu_one.md
- perf_infer_special.md
- perf_train_cpu.md
- perf_train_cpu_many.md
- perf_train_special.md
- perf_train_tpu.md
- pipeline_tutorial.md
- pr_checks.md
- preprocessing.md
- quicktour.md
- run_scripts.md
- training.md
- audio_utils.md
- file_utils.md
- generation_utils.md
- image_processing_utils.md
- modeling_utils.md
- pipelines_utils.md
- time_series_utils.md
- tokenization_utils.md
- trainer_utils.md
- callback.md
- configuration.md
- data_collator.md
- deepspeed.md
- feature_extractor.md
- image_processor.md
- logging.md
- model.md
- optimizer_schedules.md
- output.md
- pipelines.md
- processors.md
- quantization.md
- text_generation.md
- tokenizer.md
- trainer.md
- albert.md
- align.md
- altclip.md
- audio-spectrogram-transformer.md
- auto.md
- autoformer.md
- bark.md
- bart.md
- barthez.md
- bartpho.md
- beit.md
- bert-generation.md
- bert-japanese.md
- bert.md
- bertweet.md
- big_bird.md
- bigbird_pegasus.md
- biogpt.md
- bit.md
- blenderbot-small.md
- blenderbot.md
- blip-2.md
- blip.md
- bloom.md
- bridgetower.md
- bros.md
- byt5.md
- camembert.md
- canine.md
- chinese_clip.md
- clap.md
- clip.md
- clipseg.md
- clvp.md
- code_llama.md
- codegen.md
- conditional_detr.md
- convbert.md
- convnext.md
- convnextv2.md
- cpm.md
- cpmant.md
- ctrl.md
- cvt.md
- data2vec.md
- deberta-v2.md
- deberta.md
- decision_transformer.md
- deformable_detr.md
- deit.md
- deplot.md
- detr.md
- dinat.md
- asr.md
- audio_classification.md
- document_question_answering.md
- idefics.md
- image_captioning.md
- image_classification.md
- knowledge_distillation_for_image_classification.md
- language_modeling.md
- masked_language_modeling.md
- monocular_depth_estimation.md
- multiple_choice.md
- object_detection.md
- prompting.md
- question_answering.md
- semantic_segmentation.md
- summarization.md
- text-to-speech.md
- token_classification.md
- translation.md
- video_classification.md
- visual_question_answering.md
- zero_shot_image_classification.md
- zero_shot_object_detection.md
- _toctree.yml
- accelerate.md
- add_new_model.md
- attention.md
- autoclass_tutorial.md
- bertology.md
- big_models.md
- chat_templating.md
- community.md
- create_a_model.md
- custom_models.md
- fast_tokenizers.md
- generation_strategies.md
- glossary.md
- hpo_train.md
- index.md
- installation.md
- llm_tutorial.md
- model_memory_anatomy.md
- model_sharing.md
- model_summary.md
- multilingual.md
- pad_truncation.md
- peft.md
- perf_hardware.md
- perf_infer_cpu.md
- perf_infer_gpu_many.md
- perf_infer_gpu_one.md
- perf_infer_special.md
- perf_torch_compile.md
- perf_train_cpu.md
- perf_train_cpu_many.md
- perf_train_gpu_many.md
- perf_train_gpu_one.md
- perf_train_special.md
- perf_train_tpu.md
- performance.md
- perplexity.md
- philosophy.md
- pipeline_tutorial.md
- pipeline_webserver.md
- pr_checks.md
- preprocessing.md
- quicktour.md
- run_scripts.md
- serialization.md
- task_summary.md
- tasks_explained.md
- testing.md
- tokenizer_summary.md
- training.md
- troubleshooting.md
- audio_utils.md
- file_utils.md
- generation_utils.md
- image_processing_utils.md
- modeling_utils.md
- pipelines_utils.md
- time_series_utils.md
- tokenization_utils.md
- trainer_utils.md
- callback.md
- configuration.md
- data_collator.md
- feature_extractor.md
- logging.md
- model.md
- optimizer_schedules.md
- output.md
- peft.md
- pipelines.md
- processors.md
- quantization.md
- text_generation.md
- tokenizer.md
- trainer.md
- albert.md
- altclip.md
- auto.md
- autoformer.md
- bart.md
- barthez.md
- bartpho.md
- bert-japanese.md
- bert.md
- bertweet.md
- big_bird.md
- biogpt.md
- blip-2.md
- blip.md
- chameleon.md
- clip.md
- clipseg.md
- code_llama.md
- codegen.md
- cohere.md
- convbert.md
- dbrx.md
- deberta-v2.md
- deberta.md
- deepseek_v3.md
- electra.md
- encoder-decoder.md
- esm.md
- exaone4.md
- exaone4_5.md
- exaone_moe.md
- gemma.md
- gemma2.md
- gemma3.md
- gemma3n.md
- gpt2.md
- gpt_neox_japanese.md
- grounding-dino.md
- informer.md
- jamba.md
- lfm2.md
- llama.md
- llama2.md
- llama3.md
- llama4.md
- mamba.md
- mamba2.md
- marian.md
- mistral.md
- openai-gpt.md
- paligemma.md
- patchtsmixer.md
- patchtst.md
- qwen2_vl.md
- rag.md
- roberta.md
- sam_hq.md
- siglip.md
- smolvlm.md
- swin.md
- swin2sr.md
- swinv2.md
- time_series_transformer.md
- timesformer.md
- tvp.md
- vit.md
- vivit.md
- whisper.md
- xclip.md
- awq.md
- bitsandbytes.md
- eetq.md
- gptq.md
- optimum.md
- quanto.md
- quark.md
- asr.md
- audio_classification.md
- document_question_answering.md
- idefics.md
- image_captioning.md
- image_classification.md
- image_feature_extraction.md
- keypoint_detection.md
- knowledge_distillation_for_image_classification.md
- language_modeling.md
- mask_generation.md
- masked_language_modeling.md
- monocular_depth_estimation.md
- multiple_choice.md
- object_detection.md
- prompting.md
- question_answering.md
- semantic_segmentation.md
- sequence_classification.md
- summarization.md
- token_classification.md
- translation.md
- video_classification.md
- visual_question_answering.md
- zero_shot_image_classification.md
- zero_shot_object_detection.md
- _config.py
- _toctree.yml
- accelerate.md
- accelerator_selection.md
- add_new_model.md
- add_new_pipeline.md
- cache_explanation.md
- chat_extras.md
- chat_templating.md
- community.md
- contributing.md
- conversations.md
- custom_models.md
- debugging.md
- deepspeed.md
- executorch.md
- fast_tokenizers.md
- fsdp.md
- generation_strategies.md
- gguf.md
- glossary.md
- how_to_hack_models.md
- hpo_train.md
- image_processors.md
- in_translation.md
- index.md
- installation.md
- llm_optims.md
- llm_tutorial.md
- llm_tutorial_optimization.md
- model_memory_anatomy.md
- model_sharing.md
- models.md
- modular_transformers.md
- optimizers.md
- pad_truncation.md
- peft.md
- perf_hardware.md
- perf_infer_cpu.md
- perf_infer_gpu_multi.md
- perf_infer_gpu_one.md
- perf_train_cpu.md
- perf_train_cpu_many.md
- perf_train_gpu_many.md
- perf_train_gpu_one.md
- perf_train_special.md
- perplexity.md
- philosophy.md
- pipeline_gradio.md
- pipeline_tutorial.md
- pipeline_webserver.md
- pr_checks.md
- quicktour.md
- run_scripts.md
- serialization.md
- serving.md
- testing.md
- tiny_agents.md
- tokenizer_summary.md
- trainer.md
- training.md
- troubleshooting.md
- sequence_classification.md
- token_classification.md
- _config.py
- _toctree.yml
- accelerate.md
- create_a_model.md
- custom_models.md
- fast_tokenizers.md
- index.md
- installation.md
- multilingual.md
- pipeline_tutorial.md
- quicktour.md
- run_scripts.md
- training.md
- _config.py
- _toctree.yml
- add_audio_processing_components.md
- add_new_model.md
- add_new_pipeline.md
- add_vision_processing_components.md
- auto_docstring.md
- backbones.md
- contributing.md
- custom_models.md
- custom_tokenizers.md
- fast_tokenizers.md
- feature_extractors.md
- fusion_mapping.md
- how_to_hack_models.md
- image_processors.md
- index.md
- installation.md
- model_output_tracing.md
- model_sharing.md
- modeling_rules.md
- models.md
- modular_transformers.md
- monkey_patching.md
- multimodal_processing.md
- pipeline_gradio.md
- pipeline_tutorial.md
- pipeline_webserver.md
- pr_checks.md
- processors.md
- quicktour.md
- serialization.md
- testing.md
- tokenizer_summary.md
- video_processors.md
- weightconverter.md
- _config.py
- _toctree.yml
- index.md
- installation.md
- quicktour.md
- audio_utils.md
- file_utils.md
- generation_utils.md
- image_processing_utils.md
- modeling_utils.md
- pipelines_utils.md
- time_series_utils.md
- tokenization_utils.md
- trainer_utils.md
- callback.md
- configuration.md
- data_collator.md
- deepspeed.md
- feature_extractor.md
- image_processor.md
- logging.md
- model.md
- optimizer_schedules.md
- output.md
- pipelines.md
- processors.md
- quantization.md
- text_generation.md
- tokenizer.md
- trainer.md
- bert.md
- asr.md
- question_answering.md
- sequence_classification.md
- summarization.md
- token_classification.md
- translation.md
- _toctree.yml
- accelerate.md
- add_new_pipeline.md
- attention.md
- autoclass_tutorial.md
- bertology.md
- big_models.md
- chat_templating.md
- community.md
- contributing.md
- create_a_model.md
- custom_models.md
- debugging.md
- fast_tokenizers.md
- fsdp.md
- generation_strategies.md
- gguf.md
- hpo_train.md
- index.md
- installation.md
- kernels.md
- llm_tutorial.md
- model_sharing.md
- multilingual.md
- peft.md
- perf_hardware.md
- perf_infer_gpu_multi.md
- perf_torch_compile.md
- perf_train_cpu.md
- perf_train_special.md
- performance.md
- philosophy.md
- pipeline_tutorial.md
- preprocessing.md
- quicktour.md
- run_scripts.md
- serialization.md
- task_summary.md
- tiktoken.md
- tokenizer_summary.md
- training.md
- _config.py
- README.md
- TRANSLATING.md
- configuration_dummy.py
- configuration_duplicated_method.py
- configuration_my_new_model.py
- configuration_my_new_model2.py
- configuration_new_model.py
- configuration_super.py
- convert_examples.sh
- image_processing_new_imgproc_model.py
- modeling_add_function.py
- modeling_dummy_bert.py
- modeling_from_uppercase_model.py
- modeling_global_indexing.py
- modeling_multimodal2.py
- modeling_my_new_model2.py
- modeling_new_task_model.py
- modeling_roberta.py
- modeling_super.py
- modeling_switch_function.py
- modeling_test_detr.py
- modeling_test_suffix.py
- modular_add_function.py
- modular_dummy_bert.py
- modular_duplicated_method.py
- modular_from_uppercase_model.py
- modular_global_indexing.py
- modular_multimodal2.py
- modular_my_new_model.py
- modular_my_new_model2.py
- modular_new_imgproc_model.py
- modular_new_model.py
- modular_new_task_model.py
- modular_roberta.py
- modular_super.py
- modular_switch_function.py
- modular_test_detr.py
- modular_test_suffix.py
- README.md
- README.md
- requirements.txt
- run_audio_classification.py
- README.md
- requirements.txt
- run_clip.py
- README.md
- requirements.txt
- run_image_classification.py
- run_image_classification_no_trainer.py
- README.md
- requirements.txt
- run_mae.py
- run_mim.py
- run_mim_no_trainer.py
- README.md
- requirements.txt
- run_instance_segmentation.py
- run_instance_segmentation_no_trainer.py
- README.md
- requirements.txt
- run_clm.py
- run_clm_no_trainer.py
- run_fim.py
- run_fim_no_trainer.py
- run_mlm.py
- run_mlm_no_trainer.py
- run_plm.py
- README.md
- requirements.txt
- run_no_trainer.sh
- run_swag.py
- run_swag_no_trainer.py
- README.md
- requirements.txt
- run_object_detection.py
- run_object_detection_no_trainer.py
- README.md
- requirements.txt
- run_qa.py
- run_qa_beam_search.py
- run_qa_beam_search_no_trainer.py
- run_qa_no_trainer.py
- run_seq2seq_qa.py
- trainer_qa.py
- trainer_seq2seq_qa.py
- utils_qa.py
- README.md
- requirements.txt
- run_semantic_segmentation.py
- run_semantic_segmentation_no_trainer.py
- README.md
- requirements.txt
- run_wav2vec2_pretraining_no_trainer.py
- README.md
- requirements.txt
- run_speech_recognition_ctc.py
- run_speech_recognition_ctc_adapter.py
- run_speech_recognition_seq2seq.py
- README.md
- requirements.txt
- run_summarization.py
- run_summarization_no_trainer.py
- README.md
- requirements.txt
- run_classification.py
- run_glue.py
- run_glue_no_trainer.py
- run_xnli.py
- README.md
- requirements.txt
- run_generation.py
- README.md
- requirements.txt
- run.sh
- run_ner.py
- run_ner_no_trainer.py
- run_no_trainer.sh
- README.md
- requirements.txt
- run_translation.py
- run_translation_no_trainer.py
- 3d_parallel_checks.py
- _tests_requirements.txt
- conftest.py
- context_parallel.py
- continuous_batching.py
- continuous_batching_simple.py
- old_test_xla_examples.py
- README.md
- test_accelerate_examples.py
- test_pytorch_examples.py
- transformers_serve_cb_eval_job.py
- xla_spawn.py
- custom_quantization.py
- custom_quantization_int8_example.py
- README.md
- README.md
- run_greedy.py
- distributed_training.py
- 3D_parallel.py
- README.md
- run_on_remote.py
- README_ar.md
- README_bn.md
- README_de.md
- README_es.md
- README_fa.md
- README_fr.md
- README_hd.md
- README_it.md
- README_ja.md
- README_ko.md
- README_pt-br.md
- README_ro.md
- README_ru.md
- README_te.md
- README_tr.md
- README_ur.md
- README_vi.md
- README_zh-hans.md
- README_zh-hant.md
- README.md
- torch-distributed-gpu-test.py
- check_tokenizers.py
- stale.py
- __init__.py
- chat_completion.py
- completion.py
- model_manager.py
- response.py
- server.py
- transcription.py
- utils.py
- __init__.py
- add_new_model_like.py
- chat.py
- download.py
- serve.py
- system.py
- transformers.py
- __init__.py
- glue.py
- squad.py
- __init__.py
- squad_metrics.py
- __init__.py
- glue.py
- squad.py
- utils.py
- xnli.py
- __init__.py
- data_collator.py
- __init__.py
- configuration_utils.py
- fsdp.py
- mixin.py
- sharding_utils.py
- tensor_parallel.py
- utils.py
- __init__.py
- auto.py
- base.py
- configs.py
- exporter_dynamo.py
- exporter_executorch.py
- exporter_onnx.py
- utils.py
- __init__.py
- cache.py
- cache_manager.py
- cb_logits_processors.py
- continuous_api.py
- distributed.py
- initialization.py
- input_outputs.py
- model_runner.py
- offloading_manager.py
- requests.py
- scheduler.py
- utils.py
- __init__.py
- candidate_generator.py
- configuration_utils.py
- logits_process.py
- stopping_criteria.py
- streamers.py
- utils.py
- watermarking.py
- __init__.py
- configuration_utils.py
- __init__.py
- constants.py
- tokenizer.py
- __init__.py
- accelerate.py
- aqlm.py
- awq.py
- bitnet.py
- bitsandbytes.py
- compressed_tensors.py
- deepgemm.py
- deepspeed.py
- eager_paged.py
- eetq.py
- executorch.py
- fbgemm_fp8.py
- finegrained_fp8.py
- flash_attention.py
- flash_paged.py
- flex_attention.py
- fouroversix.py
- fp_quant.py
- fsdp.py
- gemma_quant.py
- ggml.py
- higgs.py
- hqq.py
- hub_kernels.py
- integration_utils.py
- liger.py
- metal_quantization.py
- moe.py
- msa_attention.py
- mxfp4.py
- neftune.py
- npu_flash_attention.py
- nvfp4.py
- peft.py
- quanto.py
- quark.py
- sdpa_attention.py
- sdpa_paged.py
- sinq.py
- sonicmoe.py
- spqr.py
- tensor_parallel.py
- tiktoken.py
- torchao.py
- tpu.py
- vptq.py
- __init__.py
- loss_d_fine.py
- loss_deformable_detr.py
- loss_deimv2.py
- loss_for_object_detection.py
- loss_grounding_dino.py
- loss_lw_detr.py
- loss_rf_detr.py
- loss_rnnt.py
- loss_rt_detr.py
- loss_tdt.py
- loss_utils.py
- __init__.py
- configuration_afmoe.py
- modeling_afmoe.py
- modular_afmoe.py
- __init__.py
- configuration_aimv2.py
- convert_aimv2_original_pytorch_to_hf.py
- modeling_aimv2.py
- modular_aimv2.py
- __init__.py
- configuration_albert.py
- convert_albert_original_tf_checkpoint_to_pytorch.py
- modeling_albert.py
- tokenization_albert.py
- __init__.py
- configuration_align.py
- convert_align_tf_to_hf.py
- modeling_align.py
- processing_align.py
- __init__.py
- configuration_altclip.py
- modeling_altclip.py
- modular_altclip.py
- processing_altclip.py
- __init__.py
- configuration_apertus.py
- modeling_apertus.py
- modular_apertus.py
- __init__.py
- configuration_arcee.py
- modeling_arcee.py
- modular_arcee.py
- __init__.py
- configuration_aria.py
- convert_aria_weights_to_hf.py
- image_processing_aria.py
- image_processing_pil_aria.py
- modeling_aria.py
- modular_aria.py
- processing_aria.py
- __init__.py
- configuration_audio_spectrogram_transformer.py
- convert_audio_spectrogram_transformer_original_to_pytorch.py
- feature_extraction_audio_spectrogram_transformer.py
- modeling_audio_spectrogram_transformer.py
- modular_audio_spectrogram_transformer.py
- __init__.py
- configuration_audioflamingo3.py
- convert_audioflamingo3_to_hf.py
- modeling_audioflamingo3.py
- modular_audioflamingo3.py
- processing_audioflamingo3.py
- __init__.py
- auto_factory.py
- auto_mappings.py
- configuration_auto.py
- feature_extraction_auto.py
- image_processing_auto.py
- modeling_auto.py
- processing_auto.py
- tokenization_auto.py
- video_processing_auto.py
- __init__.py
- configuration_autoformer.py
- modeling_autoformer.py
- __init__.py
- configuration_axk1.py
- modeling_axk1.py
- modular_axk1.py
- __init__.py
- configuration_axk2.py
- modeling_axk2.py
- modular_axk2.py
- __init__.py
- configuration_aya_vision.py
- modeling_aya_vision.py
- modular_aya_vision.py
- processing_aya_vision.py
- __init__.py
- configuration_bamba.py
- convert_mamba_ssm_checkpoint.py
- modeling_bamba.py
- modular_bamba.py
- __init__.py
- configuration_bark.py
- convert_suno_to_hf.py
- generation_configuration_bark.py
- modeling_bark.py
- processing_bark.py
- __init__.py
- configuration_bart.py
- convert_bart_original_pytorch_checkpoint_to_pytorch.py
- modeling_bart.py
- tokenization_bart.py
- __init__.py
- tokenization_barthez.py
- __init__.py
- tokenization_bartpho.py
- __init__.py
- configuration_beit.py
- convert_beit_unilm_to_pytorch.py
- image_processing_beit.py
- image_processing_pil_beit.py
- modeling_beit.py
- modular_beit.py
- __init__.py
- configuration_bert.py
- convert_bert_original_tf2_checkpoint_to_pytorch.py
- convert_bert_original_tf_checkpoint_to_pytorch.py
- convert_bert_token_dropping_original_tf2_checkpoint_to_pytorch.py
- modeling_bert.py
- tokenization_bert.py
- tokenization_bert_legacy.py
- __init__.py
- configuration_bert_generation.py
- modeling_bert_generation.py
- tokenization_bert_generation.py
- __init__.py
- tokenization_bert_japanese.py
- __init__.py
- tokenization_bertweet.py
- __init__.py
- configuration_big_bird.py
- convert_bigbird_original_tf_checkpoint_to_pytorch.py
- modeling_big_bird.py
- tokenization_big_bird.py
- __init__.py
- configuration_bigbird_pegasus.py
- convert_bigbird_pegasus_tf_to_pytorch.py
- modeling_bigbird_pegasus.py
- __init__.py
- configuration_biogpt.py
- convert_biogpt_original_pytorch_checkpoint_to_pytorch.py
- modeling_biogpt.py
- modular_biogpt.py
- tokenization_biogpt.py
- __init__.py
- configuration_bit.py
- convert_bit_to_pytorch.py
- image_processing_bit.py
- image_processing_pil_bit.py
- modeling_bit.py
- __init__.py
- configuration_bitnet.py
- modeling_bitnet.py
- modular_bitnet.py
- __init__.py
- configuration_blenderbot.py
- convert_blenderbot_original_pytorch_checkpoint_to_pytorch.py
- modeling_blenderbot.py
- tokenization_blenderbot.py
- __init__.py
- configuration_blenderbot_small.py
- modeling_blenderbot_small.py
- tokenization_blenderbot_small.py
- __init__.py
- configuration_blip.py
- convert_blip_original_pytorch_to_hf.py
- image_processing_blip.py
- image_processing_pil_blip.py
- modeling_blip.py
- modeling_blip_text.py
- processing_blip.py
- __init__.py
- configuration_blip_2.py
- convert_blip_2_original_to_pytorch.py
- modeling_blip_2.py
- processing_blip_2.py
- __init__.py
- configuration_bloom.py
- convert_bloom_original_checkpoint_to_pytorch.py
- modeling_bloom.py
- __init__.py
- configuration_blt.py
- convert_blt_weights_to_hf.py
- modeling_blt.py
- modular_blt.py
- __init__.py
- configuration_bridgetower.py
- image_processing_bridgetower.py
- image_processing_pil_bridgetower.py
- modeling_bridgetower.py
- processing_bridgetower.py
- __init__.py
- configuration_bros.py
- convert_bros_to_pytorch.py
- modeling_bros.py
- processing_bros.py
- __init__.py
- convert_byt5_original_tf_checkpoint_to_pytorch.py
- tokenization_byt5.py
- __init__.py
- configuration_camembert.py
- modeling_camembert.py
- modular_camembert.py
- tokenization_camembert.py
- __init__.py
- configuration_canine.py
- convert_canine_original_tf_checkpoint_to_pytorch.py
- modeling_canine.py
- tokenization_canine.py
- __init__.py
- configuration_chameleon.py
- convert_chameleon_weights_to_hf.py
- image_processing_chameleon.py
- image_processing_pil_chameleon.py
- modeling_chameleon.py
- processing_chameleon.py
- __init__.py
- configuration_chinese_clip.py
- convert_chinese_clip_original_pytorch_to_hf.py
- image_processing_chinese_clip.py
- image_processing_pil_chinese_clip.py
- modeling_chinese_clip.py
- modular_chinese_clip.py
- processing_chinese_clip.py
- __init__.py
- configuration_chmv2.py
- convert_chmv2_to_hf.py
- image_processing_chmv2.py
- modeling_chmv2.py
- modular_chmv2.py
- __init__.py
- configuration_clap.py
- convert_clap_original_pytorch_to_hf.py
- feature_extraction_clap.py
- modeling_clap.py
- processing_clap.py
- __init__.py
- configuration_clip.py
- convert_clip_original_pytorch_to_hf.py
- image_processing_clip.py
- image_processing_pil_clip.py
- modeling_clip.py
- processing_clip.py
- tokenization_clip.py
- __init__.py
- configuration_clipseg.py
- convert_clipseg_original_pytorch_to_hf.py
- modeling_clipseg.py
- modular_clipseg.py
- processing_clipseg.py
- __init__.py
- configuration_clvp.py
- convert_clvp_to_hf.py
- feature_extraction_clvp.py
- modeling_clvp.py
- number_normalizer.py
- processing_clvp.py
- tokenization_clvp.py
- __init__.py
- tokenization_code_llama.py
- __init__.py
- configuration_codegen.py
- modeling_codegen.py
- tokenization_codegen.py
- __init__.py
- configuration_cohere.py
- modeling_cohere.py
- modular_cohere.py
- tokenization_cohere.py
- __init__.py
- configuration_cohere2.py
- modeling_cohere2.py
- modular_cohere2.py
- __init__.py
- configuration_cohere2_moe.py
- modeling_cohere2_moe.py
- modular_cohere2_moe.py
- __init__.py
- configuration_cohere2_vision.py
- image_processing_cohere2_vision.py
- modeling_cohere2_vision.py
- modular_cohere2_vision.py
- processing_cohere2_vision.py
- __init__.py
- configuration_cohere_asr.py
- feature_extraction_cohere_asr.py
- modeling_cohere_asr.py
- modular_cohere_asr.py
- processing_cohere_asr.py
- __init__.py
- configuration_cohere_compass.py
- image_processing_cohere_compass.py
- image_processing_pil_cohere_compass.py
- modeling_cohere_compass.py
- modular_cohere_compass.py
- processing_cohere_compass.py
- video_processing_cohere_compass.py
- __init__.py
- configuration_colmodernvbert.py
- modeling_colmodernvbert.py
- modular_colmodernvbert.py
- processing_colmodernvbert.py
- __init__.py
- configuration_colpali.py
- convert_colpali_weights_to_hf.py
- modeling_colpali.py
- modular_colpali.py
- processing_colpali.py
- __init__.py
- configuration_colqwen2.py
- convert_colqwen2_weights_to_hf.py
- modeling_colqwen2.py
- modular_colqwen2.py
- processing_colqwen2.py
- __init__.py
- configuration_conditional_detr.py
- convert_conditional_detr_original_pytorch_checkpoint_to_pytorch.py
- image_processing_conditional_detr.py
- image_processing_pil_conditional_detr.py
- modeling_conditional_detr.py
- modular_conditional_detr.py
- __init__.py
- configuration_convbert.py
- convert_convbert_original_tf1_checkpoint_to_pytorch.py
- modeling_convbert.py
- tokenization_convbert.py
- __init__.py
- configuration_convnext.py
- convert_convnext_to_pytorch.py
- image_processing_convnext.py
- image_processing_pil_convnext.py
- modeling_convnext.py
- __init__.py
- configuration_convnextv2.py
- convert_convnextv2_to_pytorch.py
- modeling_convnextv2.py
- __init__.py
- configuration_cosmos3_edge.py
- image_processing_cosmos3_edge.py
- image_processing_pil_cosmos3_edge.py
- modeling_cosmos3_edge.py
- modular_cosmos3_edge.py
- processing_cosmos3_edge.py
- video_processing_cosmos3_edge.py
- __init__.py
- configuration_cosmos3_omni.py
- modeling_cosmos3_omni.py
- modular_cosmos3_omni.py
- __init__.py
- tokenization_cpm.py
- tokenization_cpm_fast.py
- __init__.py
- configuration_cpmant.py
- modeling_cpmant.py
- tokenization_cpmant.py
- __init__.py
- configuration_csm.py
- convert_csm.py
- generation_csm.py
- modeling_csm.py
- modular_csm.py
- processing_csm.py
- __init__.py
- configuration_ctrl.py
- modeling_ctrl.py
- tokenization_ctrl.py
- __init__.py
- configuration_cvt.py
- convert_cvt_original_pytorch_checkpoint_to_pytorch.py
- modeling_cvt.py
- __init__.py
- configuration_cwm.py
- modeling_cwm.py
- modular_cwm.py
- __init__.py
- configuration_d_fine.py
- convert_d_fine_original_pytorch_checkpoint_to_hf.py
- modeling_d_fine.py
- modular_d_fine.py
- __init__.py
- configuration_dab_detr.py
- convert_dab_detr_original_pytorch_checkpoint_to_pytorch.py
- modeling_dab_detr.py
- __init__.py
- configuration_dac.py
- convert_dac_checkpoint.py
- feature_extraction_dac.py
- modeling_dac.py
- __init__.py
- configuration_data2vec_audio.py
- configuration_data2vec_text.py
- configuration_data2vec_vision.py
- convert_data2vec_audio_original_pytorch_checkpoint_to_pytorch.py
- convert_data2vec_text_original_pytorch_checkpoint_to_pytorch.py
- convert_data2vec_vision_original_pytorch_checkpoint_to_pytorch.py
- modeling_data2vec_audio.py
- modeling_data2vec_text.py
- modeling_data2vec_vision.py
- modular_data2vec_audio.py
- modular_data2vec_text.py
- __init__.py
- configuration_dbrx.py
- modeling_dbrx.py
- modular_dbrx.py
- __init__.py
- configuration_deberta.py
- modeling_deberta.py
- tokenization_deberta.py
- __init__.py
- configuration_deberta_v2.py
- modeling_deberta_v2.py
- tokenization_deberta_v2.py
- __init__.py
- configuration_decision_transformer.py
- modeling_decision_transformer.py
- __init__.py
- configuration_deepseek_ocr2.py
- convert_deepseek_ocr2_weights_to_hf.py
- image_processing_deepseek_ocr2.py
- image_processing_pil_deepseek_ocr2.py
- modeling_deepseek_ocr2.py
- modular_deepseek_ocr2.py
- processing_deepseek_ocr2.py
- __init__.py
- configuration_deepseek_v2.py
- modeling_deepseek_v2.py
- modular_deepseek_v2.py
- __init__.py
- configuration_deepseek_v3.py
- modeling_deepseek_v3.py
- modular_deepseek_v3.py
- __init__.py
- configuration_deepseek_v32.py
- modeling_deepseek_v32.py
- modular_deepseek_v32.py
- __init__.py
- configuration_deepseek_v4.py
- modeling_deepseek_v4.py
- modular_deepseek_v4.py
- __init__.py
- configuration_deepseek_vl.py
- convert_deepseek_vl_weights_to_hf.py
- image_processing_deepseek_vl.py
- image_processing_pil_deepseek_vl.py
- modeling_deepseek_vl.py
- modular_deepseek_vl.py
- processing_deepseek_vl.py
- __init__.py
- configuration_deepseek_vl_hybrid.py
- convert_deepseek_vl_hybrid_weights_to_hf.py
- image_processing_deepseek_vl_hybrid.py
- image_processing_pil_deepseek_vl_hybrid.py
- modeling_deepseek_vl_hybrid.py
- modular_deepseek_vl_hybrid.py
- processing_deepseek_vl_hybrid.py
- __init__.py
- configuration_deformable_detr.py
- convert_deformable_detr_to_pytorch.py
- image_processing_deformable_detr.py
- image_processing_pil_deformable_detr.py
- modeling_deformable_detr.py
- modular_deformable_detr.py
- __init__.py
- configuration_deimv2.py
- convert_deimv2_original_pytorch_checkpoint_to_hf.py
- modeling_deimv2.py
- modular_deimv2.py
- __init__.py
- configuration_deit.py
- convert_deit_timm_to_pytorch.py
- image_processing_deit.py
- image_processing_pil_deit.py
- modeling_deit.py
- modular_deit.py
- __init__.py
- __init__.py
- configuration_depth_anything.py
- convert_depth_anything_to_hf.py
- convert_distill_any_depth_to_hf.py
- modeling_depth_anything.py
- __init__.py
- configuration_depth_pro.py
- convert_depth_pro_weights_to_hf.py
- image_processing_depth_pro.py
- modeling_depth_pro.py
- __init__.py
- configuration_detr.py
- convert_detr_original_pytorch_checkpoint_to_pytorch.py
- convert_detr_to_pytorch.py
- image_processing_detr.py
- image_processing_pil_detr.py
- modeling_detr.py
- __init__.py
- configuration_dia.py
- convert_dia_to_hf.py
- feature_extraction_dia.py
- generation_dia.py
- modeling_dia.py
- modular_dia.py
- processing_dia.py
- tokenization_dia.py
- __init__.py
- configuration_diffllama.py
- modeling_diffllama.py
- modular_diffllama.py
- __init__.py
- configuration_diffusion_gemma.py
- convert_diffusion_gemma_weights.py
- generation_diffusion_gemma.py
- modeling_diffusion_gemma.py
- modular_diffusion_gemma.py
- __init__.py
- configuration_dinat.py
- modeling_dinat.py
- __init__.py
- configuration_dinov2.py
- convert_dinov2_to_hf.py
- modeling_dinov2.py
- __init__.py
- configuration_dinov2_with_registers.py
- convert_dinov2_with_registers_to_hf.py
- modeling_dinov2_with_registers.py
- modular_dinov2_with_registers.py
- __init__.py
- configuration_dinov3_convnext.py
- convert_dinov3_convnext_to_hf.py
- modeling_dinov3_convnext.py
- __init__.py
- configuration_dinov3_vit.py
- convert_dinov3_vit_to_hf.py
- image_processing_dinov3_vit.py
- modeling_dinov3_vit.py
- modular_dinov3_vit.py
- __init__.py
- configuration_distilbert.py
- modeling_distilbert.py
- tokenization_distilbert.py
- __init__.py
- configuration_doge.py
- convert_doge_weights_to_hf.py
- modeling_doge.py
- modular_doge.py
- __init__.py
- configuration_donut_swin.py
- convert_donut_to_pytorch.py
- image_processing_donut.py
- image_processing_pil_donut.py
- modeling_donut_swin.py
- processing_donut.py
- __init__.py
- configuration_dots1.py
- modeling_dots1.py
- modular_dots1.py
- __init__.py
- configuration_dpr.py
- convert_dpr_original_checkpoint_to_pytorch.py
- modeling_dpr.py
- tokenization_dpr.py
- tokenization_dpr_fast.py
- __init__.py
- configuration_dpt.py
- convert_dinov2_depth_to_hf.py
- convert_dpt_beit_to_hf.py
- convert_dpt_hybrid_to_pytorch.py
- convert_dpt_swinv2_to_hf.py
- convert_dpt_to_pytorch.py
- image_processing_dpt.py
- image_processing_pil_dpt.py
- modeling_dpt.py
- modular_dpt.py
- __init__.py
- configuration_edgetam.py
- convert_edgetam_to_hf.py
- modeling_edgetam.py
- modular_edgetam.py
- __init__.py
- configuration_edgetam_video.py
- convert_edgetam_video_to_hf.py
- modeling_edgetam_video.py
- modular_edgetam_video.py
- __init__.py
- configuration_efficientloftr.py
- convert_efficientloftr_to_hf.py
- image_processing_efficientloftr.py
- image_processing_pil_efficientloftr.py
- modeling_efficientloftr.py
- modular_efficientloftr.py
- __init__.py
- configuration_efficientnet.py
- convert_efficientnet_to_pytorch.py
- image_processing_efficientnet.py
- image_processing_pil_efficientnet.py
- modeling_efficientnet.py
- __init__.py
- configuration_electra.py
- convert_electra_original_tf_checkpoint_to_pytorch.py
- modeling_electra.py
- __init__.py
- configuration_emu3.py
- convert_emu3_weights_to_hf.py
- image_processing_emu3.py
- modeling_emu3.py
- modular_emu3.py
- processing_emu3.py
- __init__.py
- configuration_encodec.py
- convert_encodec_checkpoint_to_pytorch.py
- feature_extraction_encodec.py
- modeling_encodec.py
- __init__.py
- configuration_encoder_decoder.py
- modeling_encoder_decoder.py
- __init__.py
- configuration_eomt.py
- convert_eomt_to_hf.py
- image_processing_eomt.py
- image_processing_pil_eomt.py
- modeling_eomt.py
- modular_eomt.py
- __init__.py
- configuration_eomt_dinov3.py
- convert_eomt_dinov3_to_hf.py
- modeling_eomt_dinov3.py
- modular_eomt_dinov3.py
- __init__.py
- configuration_ernie.py
- modeling_ernie.py
- modular_ernie.py
- __init__.py
- configuration_ernie4_5.py
- convert_ernie4_5_tokenizer.py
- modeling_ernie4_5.py
- modular_ernie4_5.py
- __init__.py
- configuration_ernie4_5_moe.py
- modeling_ernie4_5_moe.py
- modular_ernie4_5_moe.py
- __init__.py
- configuration_ernie4_5_vl_moe.py
- convert_ernie4_5_vl_moe_to_hf.py
- image_processing_ernie4_5_vl_moe.py
- image_processing_pil_ernie4_5_vl_moe.py
- modeling_ernie4_5_vl_moe.py
- modular_ernie4_5_vl_moe.py
- processing_ernie4_5_vl_moe.py
- video_processing_ernie4_5_vl_moe.py
- __init__.py
- chunk_utils.py
- data_transforms.py
- feats.py
- loss.py
- protein.py
- residue_constants.py
- rigid_utils.py
- tensor_utils.py
- __init__.py
- configuration_esm.py
- convert_esm.py
- modeling_esm.py
- modeling_esmfold.py
- tokenization_esm.py
- __init__.py
- configuration_esmc.py
- convert_esmc_checkpoint.py
- modeling_esmc.py
- modular_esmc.py
- tokenization_esmc.py
- __init__.py
- configuration_esmfold2.py
- convert_esmfold2_checkpoint.py
- generation_esmfold2.py
- modeling_esmfold2.py
- modular_esmfold2.py
- protein_utils.py
- __init__.py
- configuration_eurobert.py
- modeling_eurobert.py
- modular_eurobert.py
- __init__.py
- configuration_evolla.py
- modeling_evolla.py
- modular_evolla.py
- processing_evolla.py
- __init__.py
- configuration_exaone4.py
- modeling_exaone4.py
- modular_exaone4.py
- __init__.py
- configuration_exaone4_5.py
- modeling_exaone4_5.py
- modular_exaone4_5.py
- processing_exaone4_5.py
- __init__.py
- configuration_exaone_moe.py
- modeling_exaone_moe.py
- modular_exaone_moe.py
- __init__.py
- configuration_falcon.py
- convert_custom_code_checkpoint.py
- modeling_falcon.py
- __init__.py
- configuration_falcon_h1.py
- convert_mamba_ssm_checkpoint.py
- modeling_falcon_h1.py
- modular_falcon_h1.py
- __init__.py
- configuration_falcon_mamba.py
- modeling_falcon_mamba.py
- modular_falcon_mamba.py
- __init__.py
- configuration_fast_vlm.py
- convert_fastvlm_weights_to_hf.py
- modeling_fast_vlm.py
- modular_fast_vlm.py
- __init__.py
- configuration_fastspeech2_conformer.py
- convert_fastspeech2_conformer_original_pytorch_checkpoint_to_pytorch.py
- convert_hifigan.py
- convert_model_with_hifigan.py
- modeling_fastspeech2_conformer.py
- tokenization_fastspeech2_conformer.py
- __init__.py
- configuration_flaubert.py
- modeling_flaubert.py
- tokenization_flaubert.py
- __init__.py
- configuration_flava.py
- convert_dalle_to_flava_codebook.py
- convert_flava_original_pytorch_to_hf.py
- image_processing_flava.py
- image_processing_pil_flava.py
- modeling_flava.py
- processing_flava.py
- __init__.py
- configuration_flex_olmo.py
- modeling_flex_olmo.py
- modular_flex_olmo.py
- __init__.py
- configuration_florence2.py
- convert_florence2_original_pytorch_to_hf.py
- modeling_florence2.py
- modular_florence2.py
- processing_florence2.py
- __init__.py
- configuration_fnet.py
- convert_fnet_original_flax_checkpoint_to_pytorch.py
- modeling_fnet.py
- tokenization_fnet.py
- __init__.py
- configuration_focalnet.py
- convert_focalnet_to_hf_format.py
- modeling_focalnet.py
- __init__.py
- configuration_fsmt.py
- convert_fsmt_original_pytorch_checkpoint_to_pytorch.py
- modeling_fsmt.py
- tokenization_fsmt.py
- __init__.py
- configuration_funnel.py
- convert_funnel_original_tf_checkpoint_to_pytorch.py
- modeling_funnel.py
- tokenization_funnel.py
- __init__.py
- configuration_fuyu.py
- convert_fuyu_model_weights_to_hf.py
- image_processing_fuyu.py
- image_processing_pil_fuyu.py
- modeling_fuyu.py
- processing_fuyu.py
- __init__.py
- configuration_gemma.py
- convert_gemma_weights_to_hf.py
- modeling_gemma.py
- modular_gemma.py
- tokenization_gemma.py
- __init__.py
- configuration_gemma2.py
- convert_gemma2_weights_to_hf.py
- modeling_gemma2.py
- modular_gemma2.py
- __init__.py
- configuration_gemma3.py
- convert_gemma3_weights.py
- image_processing_gemma3.py
- image_processing_pil_gemma3.py
- modeling_gemma3.py
- modular_gemma3.py
- processing_gemma3.py
- __init__.py
- configuration_gemma3n.py
- convert_gemma3n_weights.py
- feature_extraction_gemma3n.py
- modeling_gemma3n.py
- modular_gemma3n.py
- processing_gemma3n.py
- __init__.py
- configuration_gemma4.py
- convert_gemma4_weights.py
- feature_extraction_gemma4.py
- image_processing_gemma4.py
- image_processing_pil_gemma4.py
- modeling_gemma4.py
- modular_gemma4.py
- processing_gemma4.py
- video_processing_gemma4.py
- __init__.py
- configuration_gemma4_assistant.py
- modeling_gemma4_assistant.py
- __init__.py
- configuration_gemma4_unified.py
- convert_gemma4_unified_weights.py
- feature_extraction_gemma4_unified.py
- image_processing_gemma4_unified.py
- modeling_gemma4_unified.py
- modular_gemma4_unified.py
- processing_gemma4_unified.py
- video_processing_gemma4_unified.py
- __init__.py
- configuration_gemma4_unified_assistant.py
- modeling_gemma4_unified_assistant.py
- modular_gemma4_unified_assistant.py
- __init__.py
- configuration_git.py
- convert_git_to_pytorch.py
- modeling_git.py
- processing_git.py
- __init__.py
- configuration_glm.py
- convert_glm_weights_to_hf.py
- modeling_glm.py
- modular_glm.py
- __init__.py
- configuration_glm4.py
- convert_glm4_weights_to_hf.py
- modeling_glm4.py
- modular_glm4.py
- __init__.py
- configuration_glm46v.py
- image_processing_glm46v.py
- image_processing_pil_glm46v.py
- modeling_glm46v.py
- modular_glm46v.py
- processing_glm46v.py
- video_processing_glm46v.py
- __init__.py
- configuration_glm4_moe.py
- modeling_glm4_moe.py
- modular_glm4_moe.py
- __init__.py
- configuration_glm4_moe_lite.py
- modeling_glm4_moe_lite.py
- modular_glm4_moe_lite.py
- __init__.py
- configuration_glm4v.py
- convert_glm4v_mgt_weights_to_hf.py
- image_processing_glm4v.py
- image_processing_pil_glm4v.py
- modeling_glm4v.py
- modular_glm4v.py
- processing_glm4v.py
- video_processing_glm4v.py
- __init__.py
- configuration_glm4v_moe.py
- convert_glm4v_moe_mgt_weights_to_hf.py
- modeling_glm4v_moe.py
- modular_glm4v_moe.py
- __init__.py
- configuration_glm_image.py
- image_processing_glm_image.py
- image_processing_pil_glm_image.py
- modeling_glm_image.py
- modular_glm_image.py
- processing_glm_image.py
- __init__.py
- configuration_glm_moe_dsa.py
- modeling_glm_moe_dsa.py
- modular_glm_moe_dsa.py
- __init__.py
- configuration_glm_ocr.py
- modeling_glm_ocr.py
- modular_glm_ocr.py
- __init__.py
- configuration_glmasr.py
- convert_glmasr_weights_to_hf.py
- modeling_glmasr.py
- modular_glmasr.py
- __init__.py
- __init__.py
- _typing.py
- activations.py
- audio_utils.py
- backbone_utils.py
- cache_utils.py
- configuration_utils.py
- conversion_mapping.py
- convert_slow_tokenizer.py
- convert_slow_tokenizers_checkpoints_to_fast.py
- core_model_loading.py
- debug_utils.py
- dependency_versions_check.py
- dependency_versions_table.py
- dynamic_module_utils.py
- feature_extraction_sequence_utils.py
- feature_extraction_utils.py
- file_utils.py
- fusion_mapping.py
- hf_argparser.py
- hyperparameter_search.py
- image_processing_backends.py
- image_processing_base.py
- image_processing_outputs.py
- image_processing_utils.py
- image_transforms.py
- image_utils.py
- initialization.py
- masking_utils.py
- model_debugging_utils.py
- modelcard.py
- modeling_attn_mask_utils.py
- modeling_flash_attention_utils.py
- modeling_gguf_pytorch_utils.py
- modeling_layers.py
- modeling_outputs.py
- modeling_rope_utils.py
- modeling_utils.py
- .git-blame-ignore-revs
- .gitattributes
- .gitignore
- AGENTS.md
- awesome-transformers.md
- CITATION.cff
- CLAUDE.md
- CODE_OF_CONDUCT.md
- conftest.py
- CONTRIBUTING.md
- doctest_list.txt
- ISSUES.md
- LICENSE
- Makefile
- MIGRATION_GUIDE_V5.md
- pyproject.toml
- README.md
- SECURITY.md
- setup.py
# 설치 가이드
git clone https://github.com/huggingface/transformers
깃허브에서 프로젝트 코드 전체를 내 컴퓨터로 내려받습니다.
cd transformers
방금 내려받은 프로젝트 폴더 안으로 이동합니다.
2. 공식 설치 스크립트
쉬움 추천- Python 3 pip 명령어를 쓰려면 Python이 필요합니다.
pip install "transformers[torch]"
PyPI에 배포된 패키지를 바로 설치합니다. 소스 클론이 필요 없습니다.
pip install '.[torch]'
PyPI에 배포된 패키지를 바로 설치합니다. 소스 클론이 필요 없습니다.
이 레포의 README에 적힌 실제 명령어를 그대로 가져왔습니다.
3. Docker
쉬움- Git GitHub에서 프로젝트 코드를 내려받으려면 필요합니다.
- Docker Desktop 컨테이너를 빌드하고 실행하려면 필요합니다. 설치 후 실행해서 백그라운드에 켜두세요.
docker build -f docker/transformers-all-latest-gpu/Dockerfile -t transformers .
Dockerfile을 기반으로 실행 가능한 이미지를 빌드합니다.
docker run -p 8080:80 transformers
빌드된 이미지를 실제 컨테이너로 실행합니다.
4. Python
쉬움python -m venv .my-env
파이썬 스크립트(또는 모듈)를 실행합니다.
pip install "transformers[torch]"
PyPI에 배포된 패키지를 바로 설치합니다. 소스 클론이 필요 없습니다.
uv pip install "transformers[torch]"
requirements.txt 등에 명시된 파이썬 라이브러리를 설치합니다.
pip install '.[torch]'
PyPI에 배포된 패키지를 바로 설치합니다. 소스 클론이 필요 없습니다.
uv pip install '.[torch]'
requirements.txt 등에 명시된 파이썬 라이브러리를 설치합니다.
이 레포의 README에 적힌 실제 명령어를 그대로 가져왔습니다.
5. Make
보통- Git GitHub에서 프로젝트 코드를 내려받으려면 필요합니다.
- Make Linux/macOS는 보통 기본 설치되어 있습니다. Windows는 별도 설치(예: MSYS2, WSL)가 필요합니다.
make
생성된 빌드 설정을 바탕으로 실제 컴파일을 진행해 실행 파일을 만듭니다.
