# 🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨 # This file was automatically generated from existing config files and their `model_type`s. Do NOT edit this file # manually as any edits will be overwritten by auto-generation of the file. If any change should be done, # please add the correct `cls.model_type` in your config class and run `python utils/check_auto.py --fix_and_overwrite`. # 🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨 # Copyright 2026 The HuggingFace Inc. team. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. from collections import OrderedDict CONFIG_MAPPING_NAMES = OrderedDict( [ ("afmoe", "AfmoeConfig"), ("aimv2", "Aimv2Config"), ("aimv2_text_model", "Aimv2TextConfig"), ("aimv2_vision_model", "Aimv2VisionConfig"), ("albert", "AlbertConfig"), ("align", "AlignConfig"), ("align_text_model", "AlignTextConfig"), ("align_vision_model", "AlignVisionConfig"), ("altclip", "AltCLIPConfig"), ("altclip_text_model", "AltCLIPTextConfig"), ("altclip_vision_model", "AltCLIPVisionConfig"), ("apertus", "ApertusConfig"), ("arcee", "ArceeConfig"), ("aria", "AriaConfig"), ("aria_text", "AriaTextConfig"), ("audio-spectrogram-transformer", "ASTConfig"), ("audioflamingo3", "AudioFlamingo3Config"), ("audioflamingo3_encoder", "AudioFlamingo3EncoderConfig"), ("autoformer", "AutoformerConfig"), ("aya_vision", "AyaVisionConfig"), ("bamba", "BambaConfig"), ("bark", "BarkConfig"), ("bart", "BartConfig"), ("beit", "BeitConfig"), ("bert", "BertConfig"), ("bert-generation", "BertGenerationConfig"), ("big_bird", "BigBirdConfig"), ("bigbird_pegasus", "BigBirdPegasusConfig"), ("biogpt", "BioGptConfig"), ("bit", "BitConfig"), ("bitnet", "BitNetConfig"), ("blenderbot", "BlenderbotConfig"), ("blenderbot-small", "BlenderbotSmallConfig"), ("blip", "BlipConfig"), ("blip-2", "Blip2Config"), ("blip_2_qformer", "Blip2QFormerConfig"), ("blip_2_vision_model", "Blip2VisionConfig"), ("blip_text_model", "BlipTextConfig"), ("blip_vision_model", "BlipVisionConfig"), ("bloom", "BloomConfig"), ("blt", "BltConfig"), ("blt_global_transformer", "BltGlobalTransformerConfig"), ("blt_local_decoder", "BltLocalDecoderConfig"), ("blt_local_encoder", "BltLocalEncoderConfig"), ("blt_patcher", "BltPatcherConfig"), ("bridgetower", "BridgeTowerConfig"), ("bridgetower_text_model", "BridgeTowerTextConfig"), ("bridgetower_vision_model", "BridgeTowerVisionConfig"), ("bros", "BrosConfig"), ("camembert", "CamembertConfig"), ("canine", "CanineConfig"), ("chameleon", "ChameleonConfig"), ("chameleon_vqgan", "ChameleonVQVAEConfig"), ("chinese_clip", "ChineseCLIPConfig"), ("chinese_clip_text_model", "ChineseCLIPTextConfig"), ("chinese_clip_vision_model", "ChineseCLIPVisionConfig"), ("chmv2", "CHMv2Config"), ("clap", "ClapConfig"), ("clap_audio_model", "ClapAudioConfig"), ("clap_text_model", "ClapTextConfig"), ("clip", "CLIPConfig"), ("clip_text_model", "CLIPTextConfig"), ("clip_vision_model", "CLIPVisionConfig"), ("clipseg", "CLIPSegConfig"), ("clipseg_text_model", "CLIPSegTextConfig"), ("clipseg_vision_model", "CLIPSegVisionConfig"), ("clvp", "ClvpConfig"), ("clvp_decoder", "ClvpDecoderConfig"), ("clvp_encoder", "ClvpEncoderConfig"), ("codegen", "CodeGenConfig"), ("cohere", "CohereConfig"), ("cohere2", "Cohere2Config"), ("cohere2_vision", "Cohere2VisionConfig"), ("cohere_asr", "CohereAsrConfig"), ("colmodernvbert", "ColModernVBertConfig"), ("colpali", "ColPaliConfig"), ("colqwen2", "ColQwen2Config"), ("conditional_detr", "ConditionalDetrConfig"), ("convbert", "ConvBertConfig"), ("convnext", "ConvNextConfig"), ("convnextv2", "ConvNextV2Config"), ("cpmant", "CpmAntConfig"), ("csm", "CsmConfig"), ("csm_depth_decoder_model", "CsmDepthDecoderConfig"), ("ctrl", "CTRLConfig"), ("cvt", "CvtConfig"), ("cwm", "CwmConfig"), ("d_fine", "DFineConfig"), ("dab-detr", "DabDetrConfig"), ("dac", "DacConfig"), ("data2vec-audio", "Data2VecAudioConfig"), ("data2vec-text", "Data2VecTextConfig"), ("data2vec-vision", "Data2VecVisionConfig"), ("dbrx", "DbrxConfig"), ("deberta", "DebertaConfig"), ("deberta-v2", "DebertaV2Config"), ("decision_transformer", "DecisionTransformerConfig"), ("deepseek_v2", "DeepseekV2Config"), ("deepseek_v3", "DeepseekV3Config"), ("deepseek_vl", "DeepseekVLConfig"), ("deepseek_vl_hybrid", "DeepseekVLHybridConfig"), ("deformable_detr", "DeformableDetrConfig"), ("deimv2", "Deimv2Config"), ("deit", "DeiTConfig"), ("depth_anything", "DepthAnythingConfig"), ("depth_pro", "DepthProConfig"), ("detr", "DetrConfig"), ("dia", "DiaConfig"), ("dia_decoder", "DiaDecoderConfig"), ("dia_encoder", "DiaEncoderConfig"), ("diffllama", "DiffLlamaConfig"), ("dinat", "DinatConfig"), ("dinov2", "Dinov2Config"), ("dinov2_with_registers", "Dinov2WithRegistersConfig"), ("dinov3_convnext", "DINOv3ConvNextConfig"), ("dinov3_vit", "DINOv3ViTConfig"), ("distilbert", "DistilBertConfig"), ("doge", "DogeConfig"), ("donut-swin", "DonutSwinConfig"), ("dots1", "Dots1Config"), ("dpr", "DPRConfig"), ("dpt", "DPTConfig"), ("edgetam", "EdgeTamConfig"), ("edgetam_video", "EdgeTamVideoConfig"), ("edgetam_vision_model", "EdgeTamVisionConfig"), ("efficientloftr", "EfficientLoFTRConfig"), ("efficientnet", "EfficientNetConfig"), ("electra", "ElectraConfig"), ("emu3", "Emu3Config"), ("emu3_text_model", "Emu3TextConfig"), ("emu3_vqgan", "Emu3VQVAEConfig"), ("encodec", "EncodecConfig"), ("encoder-decoder", "EncoderDecoderConfig"), ("eomt", "EomtConfig"), ("eomt_dinov3", "EomtDinov3Config"), ("ernie", "ErnieConfig"), ("ernie4_5", "Ernie4_5Config"), ("ernie4_5_moe", "Ernie4_5_MoeConfig"), ("ernie4_5_vl_moe", "Ernie4_5_VLMoeConfig"), ("ernie4_5_vl_moe_text", "Ernie4_5_VLMoeTextConfig"), ("ernie4_5_vl_moe_vision", "Ernie4_5_VLMoeVisionConfig"), ("esm", "EsmConfig"), ("eurobert", "EuroBertConfig"), ("evolla", "EvollaConfig"), ("exaone4", "Exaone4Config"), ("exaone_moe", "ExaoneMoeConfig"), ("falcon", "FalconConfig"), ("falcon_h1", "FalconH1Config"), ("falcon_mamba", "FalconMambaConfig"), ("fast_vlm", "FastVlmConfig"), ("fastspeech2_conformer", "FastSpeech2ConformerConfig"), ("fastspeech2_conformer_hifigan", "FastSpeech2ConformerHifiGanConfig"), ("fastspeech2_conformer_with_hifigan", "FastSpeech2ConformerWithHifiGanConfig"), ("flaubert", "FlaubertConfig"), ("flava", "FlavaConfig"), ("flava_image_model", "FlavaImageConfig"), ("flava_multimodal_model", "FlavaMultimodalConfig"), ("flava_text_model", "FlavaTextConfig"), ("flex_olmo", "FlexOlmoConfig"), ("florence2", "Florence2Config"), ("florence_vision", "Florence2VisionConfig"), ("fnet", "FNetConfig"), ("focalnet", "FocalNetConfig"), ("fsmt", "FSMTConfig"), ("funnel", "FunnelConfig"), ("fuyu", "FuyuConfig"), ("gemma", "GemmaConfig"), ("gemma2", "Gemma2Config"), ("gemma3", "Gemma3Config"), ("gemma3_text", "Gemma3TextConfig"), ("gemma3n", "Gemma3nConfig"), ("gemma3n_audio", "Gemma3nAudioConfig"), ("gemma3n_text", "Gemma3nTextConfig"), ("gemma3n_vision", "Gemma3nVisionConfig"), ("gemma4", "Gemma4Config"), ("gemma4_audio", "Gemma4AudioConfig"), ("gemma4_text", "Gemma4TextConfig"), ("gemma4_vision", "Gemma4VisionConfig"), ("git", "GitConfig"), ("git_vision_model", "GitVisionConfig"), ("glm", "GlmConfig"), ("glm4", "Glm4Config"), ("glm46v", "Glm46VConfig"), ("glm4_moe", "Glm4MoeConfig"), ("glm4_moe_lite", "Glm4MoeLiteConfig"), ("glm4v", "Glm4vConfig"), ("glm4v_moe", "Glm4vMoeConfig"), ("glm4v_moe_text", "Glm4vMoeTextConfig"), ("glm4v_moe_vision", "Glm4vMoeVisionConfig"), ("glm4v_text", "Glm4vTextConfig"), ("glm4v_vision", "Glm4vVisionConfig"), ("glm_image", "GlmImageConfig"), ("glm_image_text", "GlmImageTextConfig"), ("glm_image_vision", "GlmImageVisionConfig"), ("glm_image_vqmodel", "GlmImageVQVAEConfig"), ("glm_moe_dsa", "GlmMoeDsaConfig"), ("glm_ocr", "GlmOcrConfig"), ("glm_ocr_text", "GlmOcrTextConfig"), ("glm_ocr_vision", "GlmOcrVisionConfig"), ("glmasr", "GlmAsrConfig"), ("glmasr_encoder", "GlmAsrEncoderConfig"), ("glpn", "GLPNConfig"), ("got_ocr2", "GotOcr2Config"), ("gpt2", "GPT2Config"), ("gpt_bigcode", "GPTBigCodeConfig"), ("gpt_neo", "GPTNeoConfig"), ("gpt_neox", "GPTNeoXConfig"), ("gpt_neox_japanese", "GPTNeoXJapaneseConfig"), ("gpt_oss", "GptOssConfig"), ("gptj", "GPTJConfig"), ("granite", "GraniteConfig"), ("granite_speech", "GraniteSpeechConfig"), ("granite_speech_encoder", "GraniteSpeechEncoderConfig"), ("granitemoe", "GraniteMoeConfig"), ("granitemoehybrid", "GraniteMoeHybridConfig"), ("granitemoeshared", "GraniteMoeSharedConfig"), ("grounding-dino", "GroundingDinoConfig"), ("groupvit", "GroupViTConfig"), ("groupvit_text_model", "GroupViTTextConfig"), ("groupvit_vision_model", "GroupViTVisionConfig"), ("helium", "HeliumConfig"), ("hgnet_v2", "HGNetV2Config"), ("hiera", "HieraConfig"), ("higgs_audio_v2", "HiggsAudioV2Config"), ("higgs_audio_v2_tokenizer", "HiggsAudioV2TokenizerConfig"), ("hubert", "HubertConfig"), ("hunyuan_v1_dense", "HunYuanDenseV1Config"), ("hunyuan_v1_moe", "HunYuanMoEV1Config"), ("hy_v3", "HYV3Config"), ("ibert", "IBertConfig"), ("idefics", "IdeficsConfig"), ("idefics2", "Idefics2Config"), ("idefics2_perceiver", "Idefics2PerceiverConfig"), ("idefics2_vision", "Idefics2VisionConfig"), ("idefics3", "Idefics3Config"), ("idefics3_vision", "Idefics3VisionConfig"), ("idefics_perciever", "IdeficsPerceiverConfig"), ("idefics_vision", "IdeficsVisionConfig"), ("ijepa", "IJepaConfig"), ("imagegpt", "ImageGPTConfig"), ("informer", "InformerConfig"), ("instructblip", "InstructBlipConfig"), ("instructblip_qformer", "InstructBlipQFormerConfig"), ("instructblip_vision_model", "InstructBlipVisionConfig"), ("instructblipvideo", "InstructBlipVideoConfig"), ("instructblipvideo_qformer", "InstructBlipVideoQFormerConfig"), ("instructblipvideo_vision_model", "InstructBlipVideoVisionConfig"), ("internvl", "InternVLConfig"), ("internvl_vision", "InternVLVisionConfig"), ("jais2", "Jais2Config"), ("jamba", "JambaConfig"), ("janus", "JanusConfig"), ("janus_vision_model", "JanusVisionConfig"), ("janus_vqgan", "JanusVQVAEConfig"), ("jetmoe", "JetMoeConfig"), ("jina_embeddings_v3", "JinaEmbeddingsV3Config"), ("kosmos-2", "Kosmos2Config"), ("kosmos-2.5", "Kosmos2_5Config"), ("kosmos_2_5_text_model", "Kosmos2_5TextConfig"), ("kosmos_2_5_vision_model", "Kosmos2_5VisionConfig"), ("kosmos_2_text_model", "Kosmos2TextConfig"), ("kosmos_2_vision_model", "Kosmos2VisionConfig"), ("kyutai_speech_to_text", "KyutaiSpeechToTextConfig"), ("laguna", "LagunaConfig"), ("lasr_ctc", "LasrCTCConfig"), ("lasr_encoder", "LasrEncoderConfig"), ("layoutlm", "LayoutLMConfig"), ("layoutlmv2", "LayoutLMv2Config"), ("layoutlmv3", "LayoutLMv3Config"), ("layoutxlm", "LayoutXLMConfig"), ("led", "LEDConfig"), ("levit", "LevitConfig"), ("lfm2", "Lfm2Config"), ("lfm2_moe", "Lfm2MoeConfig"), ("lfm2_vl", "Lfm2VlConfig"), ("lightglue", "LightGlueConfig"), ("lighton_ocr", "LightOnOcrConfig"), ("lilt", "LiltConfig"), ("llama", "LlamaConfig"), ("llama4", "Llama4Config"), ("llama4_text", "Llama4TextConfig"), ("llama4_vision_model", "Llama4VisionConfig"), ("llava", "LlavaConfig"), ("llava_next", "LlavaNextConfig"), ("llava_next_video", "LlavaNextVideoConfig"), ("llava_onevision", "LlavaOnevisionConfig"), ("longcat_flash", "LongcatFlashConfig"), ("longformer", "LongformerConfig"), ("longt5", "LongT5Config"), ("luke", "LukeConfig"), ("lw_detr", "LwDetrConfig"), ("lw_detr_vit", "LwDetrViTConfig"), ("lxmert", "LxmertConfig"), ("m2m_100", "M2M100Config"), ("mamba", "MambaConfig"), ("mamba2", "Mamba2Config"), ("marian", "MarianConfig"), ("markuplm", "MarkupLMConfig"), ("mask2former", "Mask2FormerConfig"), ("maskformer", "MaskFormerConfig"), ("maskformer-swin", "MaskFormerSwinConfig"), ("mbart", "MBartConfig"), ("megatron-bert", "MegatronBertConfig"), ("metaclip_2", "MetaClip2Config"), ("metaclip_2_text_model", "MetaClip2TextConfig"), ("metaclip_2_vision_model", "MetaClip2VisionConfig"), ("mgp-str", "MgpstrConfig"), ("mimi", "MimiConfig"), ("minicpmv4_6", "MiniCPMV4_6Config"), ("minicpmv4_6_vision", "MiniCPMV4_6VisionConfig"), ("minimax", "MiniMaxConfig"), ("minimax_m2", "MiniMaxM2Config"), ("ministral", "MinistralConfig"), ("ministral3", "Ministral3Config"), ("mistral", "MistralConfig"), ("mistral3", "Mistral3Config"), ("mistral4", "Mistral4Config"), ("mixtral", "MixtralConfig"), ("mlcd_vision_model", "MLCDVisionConfig"), ("mllama", "MllamaConfig"), ("mllama_text_model", "MllamaTextConfig"), ("mllama_vision_model", "MllamaVisionConfig"), ("mm-grounding-dino", "MMGroundingDinoConfig"), ("mobilebert", "MobileBertConfig"), ("mobilenet_v1", "MobileNetV1Config"), ("mobilenet_v2", "MobileNetV2Config"), ("mobilevit", "MobileViTConfig"), ("mobilevitv2", "MobileViTV2Config"), ("modernbert", "ModernBertConfig"), ("modernbert-decoder", "ModernBertDecoderConfig"), ("modernvbert", "ModernVBertConfig"), ("moonshine", "MoonshineConfig"), ("moonshine_streaming", "MoonshineStreamingConfig"), ("moonshine_streaming_encoder", "MoonshineStreamingEncoderConfig"), ("moshi", "MoshiConfig"), ("moshi_depth", "MoshiDepthConfig"), ("mpnet", "MPNetConfig"), ("mpt", "MptConfig"), ("mra", "MraConfig"), ("mt5", "MT5Config"), ("musicflamingo", "MusicFlamingoConfig"), ("musicgen", "MusicgenConfig"), ("musicgen_decoder", "MusicgenDecoderConfig"), ("musicgen_melody", "MusicgenMelodyConfig"), ("musicgen_melody_decoder", "MusicgenMelodyDecoderConfig"), ("mvp", "MvpConfig"), ("nanochat", "NanoChatConfig"), ("nemotron", "NemotronConfig"), ("nemotron_h", "NemotronHConfig"), ("nllb-moe", "NllbMoeConfig"), ("nomic_bert", "NomicBertConfig"), ("nougat", "NougatConfig"), ("nystromformer", "NystromformerConfig"), ("olmo", "OlmoConfig"), ("olmo2", "Olmo2Config"), ("olmo3", "Olmo3Config"), ("olmo_hybrid", "OlmoHybridConfig"), ("olmoe", "OlmoeConfig"), ("omdet-turbo", "OmDetTurboConfig"), ("oneformer", "OneFormerConfig"), ("openai-gpt", "OpenAIGPTConfig"), ("openai_privacy_filter", "OpenAIPrivacyFilterConfig"), ("opt", "OPTConfig"), ("ovis2", "Ovis2Config"), ("owlv2", "Owlv2Config"), ("owlv2_text_model", "Owlv2TextConfig"), ("owlv2_vision_model", "Owlv2VisionConfig"), ("owlvit", "OwlViTConfig"), ("owlvit_text_model", "OwlViTTextConfig"), ("owlvit_vision_model", "OwlViTVisionConfig"), ("paddleocr_vl", "PaddleOCRVLConfig"), ("paddleocr_vl_text", "PaddleOCRTextConfig"), ("paddleocr_vl_vision", "PaddleOCRVisionConfig"), ("paligemma", "PaliGemmaConfig"), ("parakeet_ctc", "ParakeetCTCConfig"), ("parakeet_encoder", "ParakeetEncoderConfig"), ("patchtsmixer", "PatchTSMixerConfig"), ("patchtst", "PatchTSTConfig"), ("pe_audio", "PeAudioConfig"), ("pe_audio_encoder", "PeAudioEncoderConfig"), ("pe_audio_video", "PeAudioVideoConfig"), ("pe_audio_video_encoder", "PeAudioVideoEncoderConfig"), ("pe_video", "PeVideoConfig"), ("pe_video_encoder", "PeVideoEncoderConfig"), ("pegasus", "PegasusConfig"), ("pegasus_x", "PegasusXConfig"), ("perceiver", "PerceiverConfig"), ("perception_lm", "PerceptionLMConfig"), ("persimmon", "PersimmonConfig"), ("phi", "PhiConfig"), ("phi3", "Phi3Config"), ("phi4_multimodal", "Phi4MultimodalConfig"), ("phi4_multimodal_audio", "Phi4MultimodalAudioConfig"), ("phi4_multimodal_vision", "Phi4MultimodalVisionConfig"), ("phimoe", "PhimoeConfig"), ("pi0", "PI0Config"), ("pix2struct", "Pix2StructConfig"), ("pix2struct_text_model", "Pix2StructTextConfig"), ("pix2struct_vision_model", "Pix2StructVisionConfig"), ("pixio", "PixioConfig"), ("pixtral", "PixtralVisionConfig"), ("plbart", "PLBartConfig"), ("poolformer", "PoolFormerConfig"), ("pop2piano", "Pop2PianoConfig"), ("pp_chart2table", "PPChart2TableConfig"), ("pp_doclayout_v2", "PPDocLayoutV2Config"), ("pp_doclayout_v3", "PPDocLayoutV3Config"), ("pp_lcnet", "PPLCNetConfig"), ("pp_lcnet_v3", "PPLCNetV3Config"), ("pp_ocrv5_mobile_det", "PPOCRV5MobileDetConfig"), ("pp_ocrv5_mobile_rec", "PPOCRV5MobileRecConfig"), ("pp_ocrv5_server_det", "PPOCRV5ServerDetConfig"), ("pp_ocrv5_server_rec", "PPOCRV5ServerRecConfig"), ("prompt_depth_anything", "PromptDepthAnythingConfig"), ("prophetnet", "ProphetNetConfig"), ("pvt", "PvtConfig"), ("pvt_v2", "PvtV2Config"), ("qianfan_ocr", "QianfanOCRConfig"), ("qianfan_ocr_vision", "QianfanOCRVisionConfig"), ("qwen2", "Qwen2Config"), ("qwen2_5_omni", "Qwen2_5OmniConfig"), ("qwen2_5_omni_audio_encoder", "Qwen2_5OmniAudioEncoderConfig"), ("qwen2_5_omni_bigvgan", "Qwen2_5OmniBigVGANConfig"), ("qwen2_5_omni_dit", "Qwen2_5OmniDiTConfig"), ("qwen2_5_omni_talker", "Qwen2_5OmniTalkerConfig"), ("qwen2_5_omni_text", "Qwen2_5OmniTextConfig"), ("qwen2_5_omni_thinker", "Qwen2_5OmniThinkerConfig"), ("qwen2_5_omni_token2wav", "Qwen2_5OmniToken2WavConfig"), ("qwen2_5_omni_vision_encoder", "Qwen2_5OmniVisionEncoderConfig"), ("qwen2_5_vl", "Qwen2_5_VLConfig"), ("qwen2_5_vl_text", "Qwen2_5_VLTextConfig"), ("qwen2_5_vl_vision", "Qwen2_5_VLVisionConfig"), ("qwen2_audio", "Qwen2AudioConfig"), ("qwen2_audio_encoder", "Qwen2AudioEncoderConfig"), ("qwen2_moe", "Qwen2MoeConfig"), ("qwen2_vl", "Qwen2VLConfig"), ("qwen2_vl_text", "Qwen2VLTextConfig"), ("qwen2_vl_vision", "Qwen2VLVisionConfig"), ("qwen3", "Qwen3Config"), ("qwen3_5", "Qwen3_5Config"), ("qwen3_5_moe", "Qwen3_5MoeConfig"), ("qwen3_5_moe_text", "Qwen3_5MoeTextConfig"), ("qwen3_5_moe_vision", "Qwen3_5MoeVisionConfig"), ("qwen3_5_text", "Qwen3_5TextConfig"), ("qwen3_5_vision", "Qwen3_5VisionConfig"), ("qwen3_moe", "Qwen3MoeConfig"), ("qwen3_next", "Qwen3NextConfig"), ("qwen3_omni_moe", "Qwen3OmniMoeConfig"), ("qwen3_omni_moe_audio_encoder", "Qwen3OmniMoeAudioEncoderConfig"), ("qwen3_omni_moe_talker_code_predictor", "Qwen3OmniMoeTalkerCodePredictorConfig"), ("qwen3_omni_moe_talker_text", "Qwen3OmniMoeTalkerTextConfig"), ("qwen3_omni_moe_text", "Qwen3OmniMoeTextConfig"), ("qwen3_omni_moe_thinker", "Qwen3OmniMoeThinkerConfig"), ("qwen3_omni_moe_vision_encoder", "Qwen3OmniMoeVisionEncoderConfig"), ("qwen3_vl", "Qwen3VLConfig"), ("qwen3_vl_moe", "Qwen3VLMoeConfig"), ("qwen3_vl_moe_text", "Qwen3VLMoeTextConfig"), ("qwen3_vl_moe_vision", "Qwen3VLMoeVisionConfig"), ("qwen3_vl_text", "Qwen3VLTextConfig"), ("qwen3_vl_vision", "Qwen3VLVisionConfig"), ("rag", "RagConfig"), ("recurrent_gemma", "RecurrentGemmaConfig"), ("reformer", "ReformerConfig"), ("regnet", "RegNetConfig"), ("rembert", "RemBertConfig"), ("resnet", "ResNetConfig"), ("roberta", "RobertaConfig"), ("roberta-prelayernorm", "RobertaPreLayerNormConfig"), ("roc_bert", "RoCBertConfig"), ("roformer", "RoFormerConfig"), ("rt_detr", "RTDetrConfig"), ("rt_detr_resnet", "RTDetrResNetConfig"), ("rt_detr_v2", "RTDetrV2Config"), ("rwkv", "RwkvConfig"), ("sam", "SamConfig"), ("sam2", "Sam2Config"), ("sam2_hiera_det_model", "Sam2HieraDetConfig"), ("sam2_video", "Sam2VideoConfig"), ("sam2_vision_model", "Sam2VisionConfig"), ("sam3", "Sam3Config"), ("sam3_detr_decoder", "Sam3DETRDecoderConfig"), ("sam3_detr_encoder", "Sam3DETREncoderConfig"), ("sam3_geometry_encoder", "Sam3GeometryEncoderConfig"), ("sam3_lite_text", "Sam3LiteTextConfig"), ("sam3_lite_text_detr_decoder", "Sam3LiteTextDETRDecoderConfig"), ("sam3_lite_text_detr_encoder", "Sam3LiteTextDETREncoderConfig"), ("sam3_lite_text_geometry_encoder", "Sam3LiteTextGeometryEncoderConfig"), ("sam3_lite_text_mask_decoder", "Sam3LiteTextMaskDecoderConfig"), ("sam3_lite_text_text_model", "Sam3LiteTextTextConfig"), ("sam3_mask_decoder", "Sam3MaskDecoderConfig"), ("sam3_tracker", "Sam3TrackerConfig"), ("sam3_tracker_video", "Sam3TrackerVideoConfig"), ("sam3_video", "Sam3VideoConfig"), ("sam3_vision_model", "Sam3VisionConfig"), ("sam3_vit_model", "Sam3ViTConfig"), ("sam_hq", "SamHQConfig"), ("sam_hq_vision_model", "SamHQVisionConfig"), ("sam_vision_model", "SamVisionConfig"), ("seamless_m4t", "SeamlessM4TConfig"), ("seamless_m4t_v2", "SeamlessM4Tv2Config"), ("seed_oss", "SeedOssConfig"), ("segformer", "SegformerConfig"), ("seggpt", "SegGptConfig"), ("sew", "SEWConfig"), ("sew-d", "SEWDConfig"), ("shieldgemma2", "ShieldGemma2Config"), ("siglip", "SiglipConfig"), ("siglip2", "Siglip2Config"), ("siglip2_text_model", "Siglip2TextConfig"), ("siglip2_vision_model", "Siglip2VisionConfig"), ("siglip_text_model", "SiglipTextConfig"), ("siglip_vision_model", "SiglipVisionConfig"), ("slanet", "SLANetConfig"), ("slanext", "SLANeXtConfig"), ("smollm3", "SmolLM3Config"), ("smolvlm", "SmolVLMConfig"), ("smolvlm_vision", "SmolVLMVisionConfig"), ("solar_open", "SolarOpenConfig"), ("speech-encoder-decoder", "SpeechEncoderDecoderConfig"), ("speech_to_text", "Speech2TextConfig"), ("speecht5", "SpeechT5Config"), ("speecht5_hifigan", "SpeechT5HifiGanConfig"), ("splinter", "SplinterConfig"), ("squeezebert", "SqueezeBertConfig"), ("stablelm", "StableLmConfig"), ("starcoder2", "Starcoder2Config"), ("superglue", "SuperGlueConfig"), ("superpoint", "SuperPointConfig"), ("swiftformer", "SwiftFormerConfig"), ("swin", "SwinConfig"), ("swin2sr", "Swin2SRConfig"), ("swinv2", "Swinv2Config"), ("switch_transformers", "SwitchTransformersConfig"), ("t5", "T5Config"), ("t5_gemma_module", "T5GemmaModuleConfig"), ("t5gemma", "T5GemmaConfig"), ("t5gemma2", "T5Gemma2Config"), ("t5gemma2_decoder", "T5Gemma2DecoderConfig"), ("t5gemma2_encoder", "T5Gemma2EncoderConfig"), ("t5gemma2_text", "T5Gemma2TextConfig"), ("table-transformer", "TableTransformerConfig"), ("tapas", "TapasConfig"), ("textnet", "TextNetConfig"), ("time_series_transformer", "TimeSeriesTransformerConfig"), ("timesfm", "TimesFmConfig"), ("timesfm2_5", "TimesFm2_5Config"), ("timesformer", "TimesformerConfig"), ("timm_backbone", "TimmBackboneConfig"), ("timm_wrapper", "TimmWrapperConfig"), ("trocr", "TrOCRConfig"), ("tvp", "TvpConfig"), ("udop", "UdopConfig"), ("umt5", "UMT5Config"), ("unispeech", "UniSpeechConfig"), ("unispeech-sat", "UniSpeechSatConfig"), ("univnet", "UnivNetConfig"), ("upernet", "UperNetConfig"), ("uvdoc", "UVDocConfig"), ("uvdoc_backbone", "UVDocBackboneConfig"), ("vaultgemma", "VaultGemmaConfig"), ("vibevoice_acoustic_tokenizer", "VibeVoiceAcousticTokenizerConfig"), ("vibevoice_asr", "VibeVoiceAsrConfig"), ("video_llama_3", "VideoLlama3Config"), ("video_llama_3_vision", "VideoLlama3VisionConfig"), ("video_llava", "VideoLlavaConfig"), ("videomae", "VideoMAEConfig"), ("videomt", "VideomtConfig"), ("vilt", "ViltConfig"), ("vipllava", "VipLlavaConfig"), ("vision-encoder-decoder", "VisionEncoderDecoderConfig"), ("vision-text-dual-encoder", "VisionTextDualEncoderConfig"), ("visual_bert", "VisualBertConfig"), ("vit", "ViTConfig"), ("vit_mae", "ViTMAEConfig"), ("vit_msn", "ViTMSNConfig"), ("vitdet", "VitDetConfig"), ("vitmatte", "VitMatteConfig"), ("vitpose", "VitPoseConfig"), ("vitpose_backbone", "VitPoseBackboneConfig"), ("vits", "VitsConfig"), ("vivit", "VivitConfig"), ("vjepa2", "VJEPA2Config"), ("voxtral", "VoxtralConfig"), ("voxtral_encoder", "VoxtralEncoderConfig"), ("voxtral_realtime", "VoxtralRealtimeConfig"), ("voxtral_realtime_encoder", "VoxtralRealtimeEncoderConfig"), ("voxtral_realtime_text", "VoxtralRealtimeTextConfig"), ("wav2vec2", "Wav2Vec2Config"), ("wav2vec2-bert", "Wav2Vec2BertConfig"), ("wav2vec2-conformer", "Wav2Vec2ConformerConfig"), ("wavlm", "WavLMConfig"), ("whisper", "WhisperConfig"), ("xclip", "XCLIPConfig"), ("xclip_text_model", "XCLIPTextConfig"), ("xclip_vision_model", "XCLIPVisionConfig"), ("xcodec", "XcodecConfig"), ("xglm", "XGLMConfig"), ("xlm", "XLMConfig"), ("xlm-roberta", "XLMRobertaConfig"), ("xlm-roberta-xl", "XLMRobertaXLConfig"), ("xlnet", "XLNetConfig"), ("xlstm", "xLSTMConfig"), ("xmod", "XmodConfig"), ("yolos", "YolosConfig"), ("yoso", "YosoConfig"), ("youtu", "YoutuConfig"), ("zamba", "ZambaConfig"), ("zamba2", "Zamba2Config"), ("zoedepth", "ZoeDepthConfig"), ] ) SPECIAL_MODEL_TYPE_TO_MODULE_NAME = OrderedDict( [ ("aimv2_text_model", "aimv2"), ("aimv2_vision_model", "aimv2"), ("align_text_model", "align"), ("align_vision_model", "align"), ("altclip_text_model", "altclip"), ("altclip_vision_model", "altclip"), ("aria_text", "aria"), ("audio-spectrogram-transformer", "audio_spectrogram_transformer"), ("audioflamingo3_encoder", "audioflamingo3"), ("bert-generation", "bert_generation"), ("blenderbot-small", "blenderbot_small"), ("blip-2", "blip_2"), ("blip_2_qformer", "blip_2"), ("blip_2_vision_model", "blip_2"), ("blip_text_model", "blip"), ("blip_vision_model", "blip"), ("blt_global_transformer", "blt"), ("blt_local_decoder", "blt"), ("blt_local_encoder", "blt"), ("blt_patcher", "blt"), ("bridgetower_text_model", "bridgetower"), ("bridgetower_vision_model", "bridgetower"), ("chameleon_vqgan", "chameleon"), ("chinese_clip_text_model", "chinese_clip"), ("chinese_clip_vision_model", "chinese_clip"), ("clap_audio_model", "clap"), ("clap_text_model", "clap"), ("clip_text_model", "clip"), ("clip_vision_model", "clip"), ("clipseg_text_model", "clipseg"), ("clipseg_vision_model", "clipseg"), ("clvp_decoder", "clvp"), ("clvp_encoder", "clvp"), ("csm_depth_decoder_model", "csm"), ("dab-detr", "dab_detr"), ("data2vec-audio", "data2vec"), ("data2vec-text", "data2vec"), ("data2vec-vision", "data2vec"), ("deberta-v2", "deberta_v2"), ("detr", "maskformer"), ("dia_decoder", "dia"), ("dia_encoder", "dia"), ("donut-swin", "donut"), ("edgetam_vision_model", "edgetam"), ("emu3_text_model", "emu3"), ("emu3_vqgan", "emu3"), ("encoder-decoder", "encoder_decoder"), ("ernie4_5_vl_moe_text", "ernie4_5_vl_moe"), ("ernie4_5_vl_moe_vision", "ernie4_5_vl_moe"), ("fastspeech2_conformer_hifigan", "fastspeech2_conformer"), ("fastspeech2_conformer_with_hifigan", "fastspeech2_conformer"), ("flava_image_model", "flava"), ("flava_multimodal_model", "flava"), ("flava_text_model", "flava"), ("florence_vision", "florence2"), ("gemma3_text", "gemma3"), ("gemma3n_audio", "gemma3n"), ("gemma3n_text", "gemma3n"), ("gemma3n_vision", "gemma3n"), ("gemma4_audio", "gemma4"), ("gemma4_text", "gemma4"), ("gemma4_vision", "gemma4"), ("git_vision_model", "git"), ("glm4v_moe_text", "glm4v_moe"), ("glm4v_moe_vision", "glm4v_moe"), ("glm4v_text", "glm4v"), ("glm4v_vision", "glm4v"), ("glm_image_text", "glm_image"), ("glm_image_vision", "glm_image"), ("glm_image_vqmodel", "glm_image"), ("glm_ocr_text", "glm_ocr"), ("glm_ocr_vision", "glm_ocr"), ("glmasr_encoder", "glmasr"), ("granite_speech_encoder", "granite_speech"), ("grounding-dino", "grounding_dino"), ("groupvit_text_model", "groupvit"), ("groupvit_vision_model", "groupvit"), ("idefics2_perceiver", "idefics2"), ("idefics2_vision", "idefics2"), ("idefics3_vision", "idefics3"), ("idefics_perciever", "idefics"), ("idefics_vision", "idefics"), ("instructblip_qformer", "instructblip"), ("instructblip_vision_model", "instructblip"), ("instructblipvideo_qformer", "instructblipvideo"), ("instructblipvideo_vision_model", "instructblipvideo"), ("internvl_vision", "internvl"), ("janus_vision_model", "janus"), ("janus_vqgan", "janus"), ("kosmos-2", "kosmos2"), ("kosmos-2.5", "kosmos2_5"), ("kosmos_2_5_text_model", "kosmos2_5"), ("kosmos_2_5_vision_model", "kosmos2_5"), ("kosmos_2_text_model", "kosmos2"), ("kosmos_2_vision_model", "kosmos2"), ("lasr_ctc", "lasr"), ("lasr_encoder", "lasr"), ("llama4_text", "llama4"), ("llama4_vision_model", "llama4"), ("lw_detr_vit", "lw_detr"), ("maskformer-swin", "maskformer"), ("megatron-bert", "megatron_bert"), ("metaclip_2_text_model", "metaclip_2"), ("metaclip_2_vision_model", "metaclip_2"), ("mgp-str", "mgp_str"), ("minicpmv4_6_vision", "minicpmv4_6"), ("mlcd_vision_model", "mlcd"), ("mllama_text_model", "mllama"), ("mllama_vision_model", "mllama"), ("mm-grounding-dino", "mm_grounding_dino"), ("modernbert-decoder", "modernbert_decoder"), ("moonshine_streaming_encoder", "moonshine_streaming"), ("moshi_depth", "moshi"), ("musicgen_decoder", "musicgen"), ("musicgen_melody_decoder", "musicgen_melody"), ("nllb-moe", "nllb_moe"), ("omdet-turbo", "omdet_turbo"), ("openai-gpt", "openai"), ("owlv2_text_model", "owlv2"), ("owlv2_vision_model", "owlv2"), ("owlvit_text_model", "owlvit"), ("owlvit_vision_model", "owlvit"), ("paddleocr_vl_text", "paddleocr_vl"), ("paddleocr_vl_vision", "paddleocr_vl"), ("parakeet_ctc", "parakeet"), ("parakeet_encoder", "parakeet"), ("pe_audio_encoder", "pe_audio"), ("pe_audio_video_encoder", "pe_audio_video"), ("pe_video_encoder", "pe_video"), ("phi4_multimodal_audio", "phi4_multimodal"), ("phi4_multimodal_vision", "phi4_multimodal"), ("pix2struct_text_model", "pix2struct"), ("pix2struct_vision_model", "pix2struct"), ("qianfan_ocr_vision", "qianfan_ocr"), ("qwen2_5_omni_audio_encoder", "qwen2_5_omni"), ("qwen2_5_omni_bigvgan", "qwen2_5_omni"), ("qwen2_5_omni_dit", "qwen2_5_omni"), ("qwen2_5_omni_talker", "qwen2_5_omni"), ("qwen2_5_omni_text", "qwen2_5_omni"), ("qwen2_5_omni_thinker", "qwen2_5_omni"), ("qwen2_5_omni_token2wav", "qwen2_5_omni"), ("qwen2_5_omni_vision_encoder", "qwen2_5_omni"), ("qwen2_5_vl_text", "qwen2_5_vl"), ("qwen2_5_vl_vision", "qwen2_5_vl"), ("qwen2_audio_encoder", "qwen2_audio"), ("qwen2_vl_text", "qwen2_vl"), ("qwen2_vl_vision", "qwen2_vl"), ("qwen3_5_moe_text", "qwen3_5_moe"), ("qwen3_5_moe_vision", "qwen3_5_moe"), ("qwen3_5_text", "qwen3_5"), ("qwen3_5_vision", "qwen3_5"), ("qwen3_omni_moe_audio_encoder", "qwen3_omni_moe"), ("qwen3_omni_moe_talker_code_predictor", "qwen3_omni_moe"), ("qwen3_omni_moe_talker_text", "qwen3_omni_moe"), ("qwen3_omni_moe_text", "qwen3_omni_moe"), ("qwen3_omni_moe_thinker", "qwen3_omni_moe"), ("qwen3_omni_moe_vision_encoder", "qwen3_omni_moe"), ("qwen3_vl_moe_text", "qwen3_vl_moe"), ("qwen3_vl_moe_vision", "qwen3_vl_moe"), ("qwen3_vl_text", "qwen3_vl"), ("qwen3_vl_vision", "qwen3_vl"), ("roberta-prelayernorm", "roberta_prelayernorm"), ("rt_detr_resnet", "rt_detr"), ("sam2_hiera_det_model", "sam2"), ("sam2_vision_model", "sam2"), ("sam3_detr_decoder", "sam3"), ("sam3_detr_encoder", "sam3"), ("sam3_geometry_encoder", "sam3"), ("sam3_lite_text_detr_decoder", "sam3_lite_text"), ("sam3_lite_text_detr_encoder", "sam3_lite_text"), ("sam3_lite_text_geometry_encoder", "sam3_lite_text"), ("sam3_lite_text_mask_decoder", "sam3_lite_text"), ("sam3_lite_text_text_model", "sam3_lite_text"), ("sam3_mask_decoder", "sam3"), ("sam3_vision_model", "sam3"), ("sam3_vit_model", "sam3"), ("sam_hq_vision_model", "sam_hq"), ("sam_vision_model", "sam"), ("sew-d", "sew_d"), ("siglip2_text_model", "siglip2"), ("siglip2_vision_model", "siglip2"), ("siglip_text_model", "siglip"), ("siglip_vision_model", "siglip"), ("smolvlm_vision", "smolvlm"), ("speech-encoder-decoder", "speech_encoder_decoder"), ("speecht5_hifigan", "speecht5"), ("t5_gemma_module", "t5gemma"), ("t5gemma2_decoder", "t5gemma2"), ("t5gemma2_encoder", "t5gemma2"), ("t5gemma2_text", "t5gemma2"), ("table-transformer", "table_transformer"), ("unispeech-sat", "unispeech_sat"), ("uvdoc_backbone", "uvdoc"), ("video_llama_3_vision", "video_llama_3"), ("vision-encoder-decoder", "vision_encoder_decoder"), ("vision-text-dual-encoder", "vision_text_dual_encoder"), ("voxtral_encoder", "voxtral"), ("voxtral_realtime_encoder", "voxtral_realtime"), ("voxtral_realtime_text", "voxtral_realtime"), ("wav2vec2-bert", "wav2vec2_bert"), ("wav2vec2-conformer", "wav2vec2_conformer"), ("xclip", "x_clip"), ("xclip_text_model", "x_clip"), ("xclip_vision_model", "x_clip"), ("xlm-roberta", "xlm_roberta"), ("xlm-roberta-xl", "xlm_roberta_xl"), ] ) IMAGE_PROCESSOR_MAPPING_NAMES = OrderedDict( [ ("aria", {"pil": "AriaImageProcessorPil", "torchvision": "AriaImageProcessor"}), ("beit", {"pil": "BeitImageProcessorPil", "torchvision": "BeitImageProcessor"}), ("bit", {"pil": "BitImageProcessorPil", "torchvision": "BitImageProcessor"}), ("blip", {"pil": "BlipImageProcessorPil", "torchvision": "BlipImageProcessor"}), ("bridgetower", {"pil": "BridgeTowerImageProcessorPil", "torchvision": "BridgeTowerImageProcessor"}), ("chameleon", {"pil": "ChameleonImageProcessorPil", "torchvision": "ChameleonImageProcessor"}), ("chinese_clip", {"pil": "ChineseCLIPImageProcessorPil", "torchvision": "ChineseCLIPImageProcessor"}), ("chmv2", {"torchvision": "CHMv2ImageProcessor"}), ("clip", {"pil": "CLIPImageProcessorPil", "torchvision": "CLIPImageProcessor"}), ("cohere2_vision", {"torchvision": "Cohere2VisionImageProcessor"}), ( "conditional_detr", {"pil": "ConditionalDetrImageProcessorPil", "torchvision": "ConditionalDetrImageProcessor"}, ), ("convnext", {"pil": "ConvNextImageProcessorPil", "torchvision": "ConvNextImageProcessor"}), ("deepseek_vl", {"pil": "DeepseekVLImageProcessorPil", "torchvision": "DeepseekVLImageProcessor"}), ( "deepseek_vl_hybrid", {"pil": "DeepseekVLHybridImageProcessorPil", "torchvision": "DeepseekVLHybridImageProcessor"}, ), ("deformable_detr", {"pil": "DeformableDetrImageProcessorPil", "torchvision": "DeformableDetrImageProcessor"}), ("deit", {"pil": "DeiTImageProcessorPil", "torchvision": "DeiTImageProcessor"}), ("depth_pro", {"torchvision": "DepthProImageProcessor"}), ("detr", {"pil": "DetrImageProcessorPil", "torchvision": "DetrImageProcessor"}), ("dinov3_vit", {"torchvision": "DINOv3ViTImageProcessor"}), ("dpt", {"pil": "DPTImageProcessorPil", "torchvision": "DPTImageProcessor"}), ("efficientloftr", {"pil": "EfficientLoFTRImageProcessorPil", "torchvision": "EfficientLoFTRImageProcessor"}), ("efficientnet", {"pil": "EfficientNetImageProcessorPil", "torchvision": "EfficientNetImageProcessor"}), ("eomt", {"pil": "EomtImageProcessorPil", "torchvision": "EomtImageProcessor"}), ("ernie4_5_vl_moe", {"pil": "Ernie4_5_VLMoeImageProcessorPil", "torchvision": "Ernie4_5_VLMoeImageProcessor"}), ("flava", {"pil": "FlavaImageProcessorPil", "torchvision": "FlavaImageProcessor"}), ("fuyu", {"pil": "FuyuImageProcessorPil", "torchvision": "FuyuImageProcessor"}), ("gemma3", {"pil": "Gemma3ImageProcessorPil", "torchvision": "Gemma3ImageProcessor"}), ("gemma4", {"pil": "Gemma4ImageProcessorPil", "torchvision": "Gemma4ImageProcessor"}), ("glm46v", {"pil": "Glm46VImageProcessorPil", "torchvision": "Glm46VImageProcessor"}), ("glm4v", {"pil": "Glm4vImageProcessorPil", "torchvision": "Glm4vImageProcessor"}), ("glm_image", {"pil": "GlmImageImageProcessorPil", "torchvision": "GlmImageImageProcessor"}), ("glpn", {"pil": "GLPNImageProcessorPil", "torchvision": "GLPNImageProcessor"}), ("got_ocr2", {"pil": "GotOcr2ImageProcessorPil", "torchvision": "GotOcr2ImageProcessor"}), ("grounding-dino", {"pil": "GroundingDinoImageProcessorPil", "torchvision": "GroundingDinoImageProcessor"}), ("idefics", {"pil": "IdeficsImageProcessorPil", "torchvision": "IdeficsImageProcessor"}), ("idefics2", {"pil": "Idefics2ImageProcessorPil", "torchvision": "Idefics2ImageProcessor"}), ("idefics3", {"pil": "Idefics3ImageProcessorPil", "torchvision": "Idefics3ImageProcessor"}), ("imagegpt", {"pil": "ImageGPTImageProcessorPil", "torchvision": "ImageGPTImageProcessor"}), ("janus", {"pil": "JanusImageProcessorPil", "torchvision": "JanusImageProcessor"}), ("layoutlmv2", {"pil": "LayoutLMv2ImageProcessorPil", "torchvision": "LayoutLMv2ImageProcessor"}), ("layoutlmv3", {"pil": "LayoutLMv3ImageProcessorPil", "torchvision": "LayoutLMv3ImageProcessor"}), ("levit", {"pil": "LevitImageProcessorPil", "torchvision": "LevitImageProcessor"}), ("lfm2_vl", {"torchvision": "Lfm2VlImageProcessor"}), ("lightglue", {"pil": "LightGlueImageProcessorPil", "torchvision": "LightGlueImageProcessor"}), ("llama4", {"torchvision": "Llama4ImageProcessor"}), ("llava", {"pil": "LlavaImageProcessorPil", "torchvision": "LlavaImageProcessor"}), ("llava_next", {"pil": "LlavaNextImageProcessorPil", "torchvision": "LlavaNextImageProcessor"}), ("llava_onevision", {"pil": "LlavaOnevisionImageProcessorPil", "torchvision": "LlavaOnevisionImageProcessor"}), ("mask2former", {"pil": "Mask2FormerImageProcessorPil", "torchvision": "Mask2FormerImageProcessor"}), ("maskformer", {"pil": "MaskFormerImageProcessorPil", "torchvision": "MaskFormerImageProcessor"}), ("minicpmv4_6", {"pil": "MiniCPMV4_6ImageProcessorPil", "torchvision": "MiniCPMV4_6ImageProcessor"}), ("mllama", {"pil": "MllamaImageProcessorPil", "torchvision": "MllamaImageProcessor"}), ("mobilenet_v1", {"pil": "MobileNetV1ImageProcessorPil", "torchvision": "MobileNetV1ImageProcessor"}), ("mobilenet_v2", {"pil": "MobileNetV2ImageProcessorPil", "torchvision": "MobileNetV2ImageProcessor"}), ("mobilevit", {"pil": "MobileViTImageProcessorPil", "torchvision": "MobileViTImageProcessor"}), ("nougat", {"pil": "NougatImageProcessorPil", "torchvision": "NougatImageProcessor"}), ("oneformer", {"pil": "OneFormerImageProcessorPil", "torchvision": "OneFormerImageProcessor"}), ("ovis2", {"pil": "Ovis2ImageProcessorPil", "torchvision": "Ovis2ImageProcessor"}), ("owlv2", {"pil": "Owlv2ImageProcessorPil", "torchvision": "Owlv2ImageProcessor"}), ("owlvit", {"pil": "OwlViTImageProcessorPil", "torchvision": "OwlViTImageProcessor"}), ("paddleocr_vl", {"pil": "PaddleOCRVLImageProcessorPil", "torchvision": "PaddleOCRVLImageProcessor"}), ("perceiver", {"pil": "PerceiverImageProcessorPil", "torchvision": "PerceiverImageProcessor"}), ("perception_lm", {"torchvision": "PerceptionLMImageProcessor"}), ("phi4_multimodal", {"torchvision": "Phi4MultimodalImageProcessor"}), ("pi0", {"torchvision": "PI0ImageProcessor"}), ("pix2struct", {"pil": "Pix2StructImageProcessorPil", "torchvision": "Pix2StructImageProcessor"}), ("pixtral", {"pil": "PixtralImageProcessorPil", "torchvision": "PixtralImageProcessor"}), ("poolformer", {"pil": "PoolFormerImageProcessorPil", "torchvision": "PoolFormerImageProcessor"}), ("pp_chart2table", {"pil": "PPChart2TableImageProcessorPil", "torchvision": "PPChart2TableImageProcessor"}), ("pp_doclayout_v2", {"torchvision": "PPDocLayoutV2ImageProcessor"}), ("pp_doclayout_v3", {"torchvision": "PPDocLayoutV3ImageProcessor"}), ("pp_lcnet", {"torchvision": "PPLCNetImageProcessor"}), ("pp_ocrv5_server_det", {"torchvision": "PPOCRV5ServerDetImageProcessor"}), ("pp_ocrv5_server_rec", {"torchvision": "PPOCRV5ServerRecImageProcessor"}), ( "prompt_depth_anything", {"pil": "PromptDepthAnythingImageProcessorPil", "torchvision": "PromptDepthAnythingImageProcessor"}, ), ("pvt", {"pil": "PvtImageProcessorPil", "torchvision": "PvtImageProcessor"}), ("qwen2_vl", {"pil": "Qwen2VLImageProcessorPil", "torchvision": "Qwen2VLImageProcessor"}), ("rt_detr", {"pil": "RTDetrImageProcessorPil", "torchvision": "RTDetrImageProcessor"}), ("sam", {"pil": "SamImageProcessorPil", "torchvision": "SamImageProcessor"}), ("sam2", {"torchvision": "Sam2ImageProcessor"}), ("sam3", {"torchvision": "Sam3ImageProcessor"}), ("segformer", {"pil": "SegformerImageProcessorPil", "torchvision": "SegformerImageProcessor"}), ("seggpt", {"pil": "SegGptImageProcessorPil", "torchvision": "SegGptImageProcessor"}), ("siglip", {"pil": "SiglipImageProcessorPil", "torchvision": "SiglipImageProcessor"}), ("siglip2", {"pil": "Siglip2ImageProcessorPil", "torchvision": "Siglip2ImageProcessor"}), ("slanext", {"torchvision": "SLANeXtImageProcessor"}), ("smolvlm", {"pil": "SmolVLMImageProcessorPil", "torchvision": "SmolVLMImageProcessor"}), ("superglue", {"pil": "SuperGlueImageProcessorPil", "torchvision": "SuperGlueImageProcessor"}), ("superpoint", {"pil": "SuperPointImageProcessorPil", "torchvision": "SuperPointImageProcessor"}), ("swin2sr", {"pil": "Swin2SRImageProcessorPil", "torchvision": "Swin2SRImageProcessor"}), ("textnet", {"pil": "TextNetImageProcessorPil", "torchvision": "TextNetImageProcessor"}), ("tvp", {"pil": "TvpImageProcessorPil", "torchvision": "TvpImageProcessor"}), ("uvdoc", {"torchvision": "UVDocImageProcessor"}), ("video_llama_3", {"pil": "VideoLlama3ImageProcessorPil", "torchvision": "VideoLlama3ImageProcessor"}), ("videomae", {"pil": "VideoMAEImageProcessorPil", "torchvision": "VideoMAEImageProcessor"}), ("vilt", {"pil": "ViltImageProcessorPil", "torchvision": "ViltImageProcessor"}), ("vit", {"pil": "ViTImageProcessorPil", "torchvision": "ViTImageProcessor"}), ("vitmatte", {"pil": "VitMatteImageProcessorPil", "torchvision": "VitMatteImageProcessor"}), ("vitpose", {"pil": "VitPoseImageProcessorPil", "torchvision": "VitPoseImageProcessor"}), ("yolos", {"pil": "YolosImageProcessorPil", "torchvision": "YolosImageProcessor"}), ("zoedepth", {"pil": "ZoeDepthImageProcessorPil", "torchvision": "ZoeDepthImageProcessor"}), ] ) VIDEO_PROCESSOR_MAPPING_NAMES = OrderedDict( [ ("ernie4_5_vl_moe", "Ernie4_5_VLMoeVideoProcessor"), ("gemma4", "Gemma4VideoProcessor"), ("glm46v", "Glm46VVideoProcessor"), ("glm4v", "Glm4vVideoProcessor"), ("instructblipvideo", "InstructBlipVideoVideoProcessor"), ("internvl", "InternVLVideoProcessor"), ("llava_next_video", "LlavaNextVideoVideoProcessor"), ("llava_onevision", "LlavaOnevisionVideoProcessor"), ("minicpmv4_6", "MiniCPMV4_6VideoProcessor"), ("pe_video", "PeVideoVideoProcessor"), ("perception_lm", "PerceptionLMVideoProcessor"), ("qwen2_vl", "Qwen2VLVideoProcessor"), ("qwen3_vl", "Qwen3VLVideoProcessor"), ("sam2_video", "Sam2VideoVideoProcessor"), ("smolvlm", "SmolVLMVideoProcessor"), ("video_llama_3", "VideoLlama3VideoProcessor"), ("video_llava", "VideoLlavaVideoProcessor"), ("videomae", "VideoMAEVideoProcessor"), ("videomt", "VideomtVideoProcessor"), ("vjepa2", "VJEPA2VideoProcessor"), ] )