tunglinwood 42fadebecb [Model] Add support for moonshotai/Kimi-Audio-7B-Instruct (#36127) ...

Signed-off-by: tunglinwood <tunglinwood@gmail.com>
Signed-off-by: tunglinwood <tomwu.tunglin@gmail.com>
Signed-off-by: tunglinwood <113751333+tunglinwood@users.noreply.github.com>

2026-03-10 21:24:48 -07:00

..

transformers

Fix import that was moved in Transformers 5.2.0 (#36120)

2026-03-05 13:59:44 +00:00

__init__.py

…

adapters.py

[Bugfix] Fix score layer quantization for sequence classification models - Qwen3 (VL) Reranker (#35849)

2026-03-04 20:57:20 -08:00

afmoe.py

…

aimv2.py

…

apertus.py

…

arcee.py

…

arctic.py

…

aria.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

audioflamingo3.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

AXK1.py

[Model] Add huggingface skt/A.X-K1 model (#32407)

2026-02-27 09:26:02 -08:00

aya_vision.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

bagel.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

baichuan.py

…

bailing_moe_linear.py

[Model] Ring 2.5 (#35102)

2026-02-26 02:17:11 -08:00

bailing_moe.py

…

bamba.py

…

bee.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

bert_with_rope.py

…

bert.py

…

blip2.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

blip.py

…

bloom.py

…

chameleon.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

chatglm.py

…

clip.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

cohere2_vision.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

colbert.py

[Model] Consolidate score logic by introduce score_type (#36479)

2026-03-10 13:32:25 +00:00

colmodernvbert.py

[Model] Consolidate score logic by introduce score_type (#36479)

2026-03-10 13:32:25 +00:00

colqwen3.py

[Model] Consolidate score logic by introduce score_type (#36479)

2026-03-10 13:32:25 +00:00

commandr.py

…

config.py

Reapply [Attention] Refactor check_and_update_config (#35122)

2026-03-09 07:17:14 -07:00

dbrx.py

…

deepencoder2.py

…

deepencoder.py

[PluggableLayer][MM] Add PluggableLayer for RelPosAttention (#33753)

2026-03-03 19:41:27 -08:00

deepseek_eagle.py

…

deepseek_mtp.py

[MTP] Validate that MTP weights are actually loaded (#35548)

2026-02-28 12:27:40 +08:00

deepseek_ocr2.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

deepseek_ocr.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

deepseek_v2.py

[Bugfix] Fix compressed-tensors quantization failure for DeepSeek-R1 on MI300x (#36247)

2026-03-07 09:26:59 +00:00

deepseek_vl2.py

[Misc] Lazy import registered processors (#36024)

2026-03-06 06:06:45 -08:00

dots1.py

…

dots_ocr.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

eagle2_5_vl.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

ernie45_moe.py

Remove padding_index from models that don't use it for better Transformers v5 compatibility (#35189)

2026-02-24 08:04:46 -08:00

ernie45_vl_moe.py

[Bugfix][CI] fix typos (#34934)

2026-03-05 17:05:46 +00:00

ernie45_vl.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

ernie45.py

…

ernie_mtp.py

…

exaone4.py

…

exaone_moe_mtp.py

…

exaone_moe.py

…

exaone.py

…

extract_hidden_states.py

[Spec Decode] Add hidden states extraction system (#33736)

2026-03-02 14:29:09 -05:00

fairseq2_llama.py

…

falcon_h1.py

…

falcon.py

…

fireredasr2.py

[Misc] Lazy import registered processors (#36024)

2026-03-06 06:06:45 -08:00

flex_olmo.py

…

funasr.py

FunASR model bugfix (#36633)

2026-03-10 08:14:21 -07:00

funaudiochat.py

[Chore] Remove debug code in model implementation (#35883)

2026-03-03 19:50:58 -08:00

fuyu.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

gemma2.py

…

gemma3_mm.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

gemma3.py

…

gemma3n_audio_utils.py

…

gemma3n_mm.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

gemma3n.py

…

gemma.py

…

glm4_1v.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

glm4_moe_lite_mtp.py

…

glm4_moe_lite.py

…

glm4_moe_mtp.py

…

glm4_moe.py

…

glm4.py

…

glm4v.py

[Misc] Move processors to transformers_utils (#35953)

2026-03-09 11:31:39 +08:00

glm_ocr_mtp.py

…

glm_ocr.py

…

glm.py

…

glmasr_utils.py

[Refactor] Remove dead private func _fp8_perm and _extract_mask_for_item (#35068)

2026-02-23 05:05:20 -08:00

glmasr.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

gpt2.py

…

gpt_bigcode.py

…

gpt_j.py

…

gpt_neox.py

…

gpt_oss.py

[Bugfix][Model] Fix gpt-oss batch invariance (#35404)

2026-02-27 20:41:24 +00:00

granite_speech.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

granite.py

…

granitemoe.py

…

granitemoehybrid.py

fix: Apply embedding_multiplier to inputs_embeds (#34813)

2026-02-23 00:42:46 -08:00

granitemoeshared.py

Remove padding_index from models that don't use it for better Transformers v5 compatibility (#35189)

2026-02-24 08:04:46 -08:00

gritlm.py

…

grok1.py

Remove padding_index from models that don't use it for better Transformers v5 compatibility (#35189)

2026-02-24 08:04:46 -08:00

h2ovl.py

[Refactor] Decouple TimingContext from InputProcessingContext (#35083)

2026-02-23 14:15:50 +00:00

hunyuan_v1.py

Remove padding_index from models that don't use it for better Transformers v5 compatibility (#35189)

2026-02-24 08:04:46 -08:00

hunyuan_vision.py

[Misc] Standardize handling of mm_processor_kwargs.size (#35284)

2026-02-26 13:05:46 +00:00

hyperclovax_vision_v2.py

[Model] Add HyperCLOVAX-SEED-Think-32B vision-language model support (#31471)

2026-03-10 10:59:19 +08:00

hyperclovax_vision.py

[Model] Add HyperCLOVAX-SEED-Think-32B vision-language model support (#31471)

2026-03-10 10:59:19 +08:00

idefics2_vision_model.py

[CI][MM] Gate vision encoder attention mask to MiniCPM only, fixing Aria regression (#36206)

2026-03-06 01:17:08 -08:00

idefics3.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

interfaces_base.py

[Model] Consolidate score logic by introduce score_type (#36479)

2026-03-10 13:32:25 +00:00

interfaces.py

[Model] Consolidate score logic by introduce score_type (#36479)

2026-03-10 13:32:25 +00:00

intern_vit.py

…

internlm2_ve.py

…

internlm2.py

…

interns1_pro.py

[CI/Build] Enable tests for recent day-0 new models (#34585)

2026-02-15 18:17:04 -08:00

interns1_vit.py

…

interns1.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

internvl.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

iquest_loopcoder.py

…

isaac.py

[Bugfix][CI] fix typos (#34934)

2026-03-05 17:05:46 +00:00

jais2.py

Remove padding_index from models that don't use it for better Transformers v5 compatibility (#35189)

2026-02-24 08:04:46 -08:00

jais.py

…

jamba.py

…

jina_vl.py

…

kanana_v.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

keye_vl1_5.py

…

keye.py

[Bugfix][CI] fix typos (#34934)

2026-03-05 17:05:46 +00:00

kimi_audio.py

[Model] Add support for moonshotai/Kimi-Audio-7B-Instruct (#36127)

2026-03-10 21:24:48 -07:00

kimi_k25_vit.py

[Bugfix] Add quant_config in ViT of Kimi-K2.5 (#34501)

2026-02-13 16:05:34 +00:00

kimi_k25.py

[Security] Respect user trust_remote_code setting in NemotronVL and KimiK25 (#36192)

2026-03-05 22:15:19 -08:00

kimi_linear.py

Remove padding_index from models that don't use it for better Transformers v5 compatibility (#35189)

2026-02-24 08:04:46 -08:00

kimi_vl.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

lfm2_moe.py

[Models] LFM2: Support LoRA (#34921)

2026-02-19 22:07:23 -08:00

lfm2_siglip2.py

[BE] Rename should_torch_compile_mm_vit to should_torch_compile_mm_encoder (#36281)

2026-03-09 18:22:05 +00:00

lfm2_vl.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

lfm2.py

[Models] LFM2: Support LoRA (#34921)

2026-02-19 22:07:23 -08:00

lightonocr.py

…

llama4_eagle.py

[Spec Decode] Reduce TP communication for speculative decoding draft token generation (#34049)

2026-02-22 14:59:16 -08:00

llama4.py

[Llama4,CI] Bring back Llama-4 bug fixes, and also fix Maverick tests (#35033)

2026-02-23 09:04:34 -05:00

llama_eagle3.py

…

llama_eagle.py

…

llama.py

…

llava_next_video.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

llava_next.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

llava_onevision.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

llava.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

longcat_flash_mtp.py

…

longcat_flash.py

[Bugfix][CI] fix typos (#34934)

2026-03-05 17:05:46 +00:00

mamba2.py

[Model][Spec Decode] Nemotron-H MTP and Mamba Speculative Decoding Support (#33726)

2026-02-24 09:49:56 -08:00

mamba.py

…

medusa.py

…

midashenglm.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

mimo_mtp.py

…

mimo_v2_flash.py

…

mimo.py

…

minicpm3.py

…

minicpm_eagle.py

…

minicpm.py

[Model Bash][DeepSeekR1] Remove Shared Expert Clone (#34344)

2026-02-19 07:56:14 -08:00

minicpmo.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

minicpmv.py

[CI][MM] Gate vision encoder attention mask to MiniCPM only, fixing Aria regression (#36206)

2026-03-06 01:17:08 -08:00

minimax_m2.py

…

minimax_text_01.py

Remove padding_index from models that don't use it for better Transformers v5 compatibility (#35189)

2026-02-24 08:04:46 -08:00

minimax_vl_01.py

…

mistral3.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

mistral_large_3_eagle.py

…

mistral_large_3.py

…

mistral.py

…

mixtral.py

…

mllama4.py

[BE] Rename should_torch_compile_mm_vit to should_torch_compile_mm_encoder (#36281)

2026-03-09 18:22:05 +00:00

mlp_speculator.py

…

modernbert.py

…

module_mapping.py

…

molmo2.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

molmo.py

[Misc] Move processors to transformers_utils (#35953)

2026-03-09 11:31:39 +08:00

moonvit.py

…

mpt.py

…

musicflamingo.py

…

nano_nemotron_vl.py

[Model] Nano Nemotron VL - fast media preprocessing (#35657)

2026-03-08 03:04:05 -07:00

nemotron_h_mtp.py

[Model][Spec Decode] Nemotron-H MTP and Mamba Speculative Decoding Support (#33726)

2026-02-24 09:49:56 -08:00

nemotron_h.py

[Bugfix][CI] fix typos (#34934)

2026-03-05 17:05:46 +00:00

nemotron_nas.py

Remove padding_index from models that don't use it for better Transformers v5 compatibility (#35189)

2026-02-24 08:04:46 -08:00

nemotron_parse.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

nemotron_vl.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

nemotron.py

…

nvlm_d.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

olmo2.py

…

olmo_hybrid.py

[Model] Add support for OLMo Hybrid (#32550)

2026-03-05 14:51:06 -05:00

olmo.py

…

olmoe.py

…

opencua.py

…

openpangu_mtp.py

…

openpangu_vl.py

…

openpangu.py

Remove padding_index from models that don't use it for better Transformers v5 compatibility (#35189)

2026-02-24 08:04:46 -08:00

opt.py

…

orion.py

…

ouro.py

…

ovis2_5.py

[Deprecation] Deprecate code in 0.17 as scheduled (#35441)

2026-02-28 17:32:37 +00:00

ovis.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

paddleocr_vl.py

[Bugfix][CI] fix typos (#34934)

2026-03-05 17:05:46 +00:00

paligemma.py

[Refactor] Decouple TimingContext from InputProcessingContext (#35083)

2026-02-23 14:15:50 +00:00

parakeet.py

ParakeetProjection.norm = RMSNorm instead of nn.LayerNorm (#36133)

2026-03-05 17:29:30 +00:00

persimmon.py

…

phi3.py

…

phi3v.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

phi4mm_audio.py

[Bugfix][CI] fix typos (#34934)

2026-03-05 17:05:46 +00:00

phi4mm_utils.py

[Bugfix][CI] fix typos (#34934)

2026-03-05 17:05:46 +00:00

phi4mm.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

phi.py

…

phimoe.py

…

pixtral.py

[Misc] Move processors to transformers_utils (#35953)

2026-03-09 11:31:39 +08:00

plamo2.py

[Model][Spec Decode] Nemotron-H MTP and Mamba Speculative Decoding Support (#33726)

2026-02-24 09:49:56 -08:00

plamo3.py

Remove padding_index from models that don't use it for better Transformers v5 compatibility (#35189)

2026-02-24 08:04:46 -08:00

qwen2_5_omni_thinker.py

Support online use_audio_in_video (#36319)

2026-03-09 07:16:44 -07:00

qwen2_5_vl.py

[BE] Rename should_torch_compile_mm_vit to should_torch_compile_mm_encoder (#36281)

2026-03-09 18:22:05 +00:00

qwen2_audio.py

[BUGFIX]Fix Qwen-Omni models audio max_token_per_item estimation error leading to encoder_cache_size is 0 (#35994)

2026-03-05 09:16:29 -08:00

qwen2_moe.py

…

qwen2_rm.py

…

qwen2_vl.py

[Bugfix][CI] fix typos (#34934)

2026-03-05 17:05:46 +00:00

qwen2.py

…

qwen3_5_mtp.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

qwen3_5.py

[Bugfix] Fix Qwen3-Next in_proj_ba weight sharding with TP > 1 (#36242)

2026-03-09 19:16:26 -07:00

qwen3_asr_realtime.py

[Bugfix] Add missing dynamic_arg_dims for Qwen3-ASR torch.compile (#35869)

2026-03-04 08:29:01 +00:00

qwen3_asr.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

qwen3_moe.py

[Bugfix][Model] Fix FP8 k_scale/v_scale not loaded for Qwen3-MoE (#35656)

2026-03-04 13:15:38 +00:00

qwen3_next_mtp.py

…

qwen3_next.py

[Bugfix] Fix Qwen3-Next in_proj_ba weight sharding with TP > 1 (#36242)

2026-03-09 19:16:26 -07:00

qwen3_omni_moe_thinker.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

qwen3_vl_moe.py

[Bugfix][Model] Fix FP8 k_scale/v_scale not loaded for Qwen3-MoE (#35656)

2026-03-04 13:15:38 +00:00

qwen3_vl.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

qwen3.py

…

qwen_vl.py

[Misc] Move processors to transformers_utils (#35953)

2026-03-09 11:31:39 +08:00

qwen.py

…

radio.py

Use MMEncoderAttention (=use FlashAttention) instead of torch.sdpa in radio.py (#35653)

2026-03-04 08:43:13 -08:00

registry.py

[Model] Add support for moonshotai/Kimi-Audio-7B-Instruct (#36127)

2026-03-10 21:24:48 -07:00

roberta.py

…

rvl.py

[Refactor] Simplify dummy data generation (#35025)

2026-02-22 20:55:27 -08:00

sarvam.py

Adding support to Sarvam's MoE models (#33942)

2026-03-08 01:16:24 +08:00

seed_oss.py

…

siglip2navit.py

…

siglip.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

skyworkr1v.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

smolvlm.py

…

solar.py

…

stablelm.py

…

starcoder2.py

…

step1.py

…

step3_text.py

…

step3_vl.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

step3p5_mtp.py

…

step3p5.py

[Bugfix][CI] fix typos (#34934)

2026-03-05 17:05:46 +00:00

step_vl.py

…

tarsier.py

…

telechat2.py

…

teleflm.py

…

terratorch.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

ultravox.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

utils.py

[Transformers backend] Ignore MTP weights when num_nextn_predict_layers=0 (#34888)

2026-02-27 19:39:23 +00:00

vision.py

[BE] Rename should_torch_compile_mm_vit to should_torch_compile_mm_encoder (#36281)

2026-03-09 18:22:05 +00:00

voxtral_realtime.py

[Misc] Move processors to transformers_utils (#35953)

2026-03-09 11:31:39 +08:00

voxtral.py

[Misc] Move processors to transformers_utils (#35953)

2026-03-09 11:31:39 +08:00

voyage.py

…

whisper_causal.py

Reapply [Attention] Refactor check_and_update_config (#35122)

2026-03-09 07:17:14 -07:00

whisper_utils.py

…

whisper.py

Increase Flexibility for OOV Multimodal Token Handling (#34858)

2026-03-08 20:30:49 -07:00

zamba2.py

…