3 people c7ca0bccae [ROCm][Perf] Add Fused Shared Expert (FSE) support for GLM-4.5/6/7 (#44313) ...

Signed-off-by: Olga Miroshnichenko <olga.miroshnichenko@amd.com>
Signed-off-by: Mehdi Ghanimifard <mehdi.ghanimifard@amd.com>
Co-authored-by: Mehdi Ghanimifard <mghanimi@amd.com>
Co-authored-by: Mehdi Ghanimifard <mehdi.ghanimifard@amd.com>

2026-06-28 00:04:08 -07:00

..

transformers

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

__init__.py

…

adapters.py

[Bugfix][Pooling] Fix silent weight corruption with buffer-reusing iterators (#39650)

2026-04-13 19:37:27 +00:00

afmoe.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

aimv2.py

[MM] Align the prefix of MMEncoderAttention with Attention (#33750)

2026-02-04 04:07:30 +00:00

apertus.py

[Migration] Migrate GGUF quantization support to plugin (#39612)

2026-06-12 12:02:21 -07:00

arcee.py

Remove KV cache scale boilerplate from model weight loading methods (#43167)

2026-06-05 05:19:04 -07:00

arctic.py

[MoE] Remove inplace fused experts mechanism (#43727)

2026-05-27 20:00:19 -07:00

aria.py

[Bugfix] Fix weight loading issues caused by #41184 (#45054)

2026-06-10 01:20:13 -04:00

audioflamingo3.py

[ROCm][CI] Gate incompatible HF references on Transformers v5 (#41532)

2026-06-16 20:34:11 +08:00

AXK1.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

aya_vision.py

[BugFix] wire make_empty_intermediate_tensors on AyaVision and Voxtral (#43118)

2026-05-22 05:26:41 -07:00

bagel.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

bailing_moe_linear.py

[MoE Refactor] FusedMoE/MoERunner inversion refactor (#41184)

2026-06-08 10:42:58 -04:00

bailing_moe.py

[Bugfix] fix wrong partial_rotary_factor calculation for bailing_moe model. (#43770)

2026-06-01 02:42:49 -07:00

bee.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

bert_with_rope.py

[MoE] Remove inplace fused experts mechanism (#43727)

2026-05-27 20:00:19 -07:00

bert.py

[Perf][4/n] Eliminate various GPU<->CPU syncs (#42347)

2026-05-19 10:35:54 -04:00

blip2.py

[Model] Fix missing maybe_prefix (#42280)

2026-05-11 09:04:06 +00:00

blip.py

[MM] Align the prefix of MMEncoderAttention with Attention (#33750)

2026-02-04 04:07:30 +00:00

bloom.py

[5/N][Attention] Finish eliminating vllm/attention folder (#32064)

2026-01-27 10:02:51 -05:00

chameleon.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

chatglm.py

[Misc] Cleanup more configs and processors (#37560)

2026-03-19 15:45:23 +00:00

cheers.py

[Model] Add support for Cheers multimodal model (#38788)

2026-04-02 21:01:40 +08:00

clip.py

[Misc][MM] Remove redundant code in CLIPAttention (#43046)

2026-05-19 08:43:16 +00:00

cohere2_moe.py

[Cohere] Fix Cohere2MoE weight loading when using Transformers ≥5.10 (#44747)

2026-06-09 06:27:40 -07:00

cohere2_vision.py

fix(quantization): Fix AWQ dequantize on Intel XPU and refactor AutoAWQ config (#42727)

2026-06-18 20:12:28 +08:00

cohere_asr.py

[Model] Fix missing maybe_prefix (#42280)

2026-05-11 09:04:06 +00:00

cohere_eagle.py

Remove KV cache scale boilerplate from model weight loading methods (#43167)

2026-06-05 05:19:04 -07:00

colbert.py

Model/colbert autoweightsloader (#44999)

2026-06-09 23:58:50 -07:00

colmodernvbert.py

[Model] Update ColModernVBERT to support latest HF checkpoint (#39307)

2026-04-09 10:48:51 +08:00

colpali.py

[Model] Add ColPali late interaction model for multi-modal retrieval (#36818)

2026-03-13 02:18:57 +00:00

colqwen3_5.py

[Model] ColQwen3.5: fix retrieval correctness (bias + bidirectional) (#46108)

2026-06-22 17:25:54 +08:00

colqwen3.py

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

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

commandr.py

[Bugfix] CohereModel.load_weights: skip modelopt _quantizer.* keys (#43495)

2026-06-11 08:14:21 -07:00

config.py

[Model] Support Unlimited OCR (#46564)

2026-06-27 23:09:18 -07:00

conformer_encoder.py

[model] support FireRedLID (#39290)

2026-04-10 08:43:58 +00:00

cosmos3.py

Add Cosmos3 Reasoner model (#43356)

2026-05-28 09:43:55 -07:00

dbrx.py

[MoE Refactor] FusedMoE/MoERunner inversion refactor (#41184)

2026-06-08 10:42:58 -04:00

deepencoder2.py

[PluggableLayer][MM] Add PluggableLayer for CustomQwen2Decoder (#37293)

2026-03-20 06:24:07 +00:00

deepencoder.py

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

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

deepseek_eagle3.py

[SpecDecode] Reduce TP communication for large-vocab draft models speculative decoding (#39419)

2026-06-10 07:59:24 +00:00

deepseek_eagle.py

[Model] Fix missing maybe_prefix (#42280)

2026-05-11 09:04:06 +00:00

deepseek_mtp.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

deepseek_ocr2.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

deepseek_ocr.py

[Model] Support Unlimited OCR (#46564)

2026-06-27 23:09:18 -07:00

deepseek_v2.py

[Model] Support Unlimited OCR (#46564)

2026-06-27 23:09:18 -07:00

deepseek_vl2.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

diffusion_gemma.py

[ModelRunner V2] Support realtime embeddings (#46762)

2026-06-26 19:42:27 -07:00

dots_ocr.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

eagle2_5_vl.py

[2/3] Refactor InternVL-based processors (#37324)

2026-03-18 22:22:19 +08:00

ernie45_moe.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

ernie45_vl_moe.py

[MoE Refactor] FusedMoE/MoERunner inversion refactor (#41184)

2026-06-08 10:42:58 -04:00

ernie45_vl.py

[Model] Use mm_features for Ernie-4.5 VL M-RoPE (#39753)

2026-04-14 01:11:52 -07:00

ernie45.py

…

ernie_mtp.py

[Chore] Update type annotation of input_ids in model forward (#33063)

2026-01-26 06:02:10 -08:00

exaone4_5_mtp.py

Fix EXAONE-4.5 to align with Transformers update (#42246)

2026-05-11 10:25:31 +00:00

exaone4_5.py

[Bugfix] Fix missing sequence_lengths in EXAONE-4.5 vision encoder (#45073)

2026-06-10 15:44:34 +01:00

exaone4.py

[Migration] Migrate GGUF quantization support to plugin (#39612)

2026-06-12 12:02:21 -07:00

exaone_moe_mtp.py

Add EXAONE-4.5 (#39388)

2026-04-09 20:53:26 -07:00

exaone_moe.py

[Model] Remove InternLMForCausalLM registry alias (#45128)

2026-06-11 20:02:51 +00:00

exaone.py

[Migration] Migrate GGUF quantization support to plugin (#39612)

2026-06-12 12:02:21 -07:00

extract_hidden_states.py

[Spec Decode] Support hybrid attention models in extract_hidden_states (#39949)

2026-05-13 10:45:53 -07:00

fairseq2_llama.py

…

falcon_h1.py

[Model] Use AutoWeightsLoader for FalconH1 (#39092)

2026-04-07 16:25:17 +08:00

falcon.py

[Misc] Cleanup more configs and processors (#37560)

2026-03-19 15:45:23 +00:00

fireredasr2.py

[Audio] Bundle get_generation_prompt() params into SpeechToTextParams (#36268)

2026-04-22 12:24:18 +08:00

fireredlid.py

[model] support FireRedLID (#39290)

2026-04-10 08:43:58 +00:00

flex_olmo.py

[MoE Refactor] Rename FusedMoE.make_expert_params_mapping to fused_moe_make_expert_params_mapping (#40671)

2026-04-23 11:22:34 -04:00

funasr.py

[Bugfix] Fix FunASR-Nano crash during initialization (#44215)

2026-06-07 20:00:02 -07:00

funaudiochat.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

fuyu.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

gemma2.py

Remove KV cache scale boilerplate from model weight loading methods (#43167)

2026-06-05 05:19:04 -07:00

gemma3_mm.py

[Perf][4/n] Eliminate various GPU<->CPU syncs (#42347)

2026-05-19 10:35:54 -04:00

gemma3.py

[Migration] Migrate GGUF quantization support to plugin (#39612)

2026-06-12 12:02:21 -07:00

gemma3n_audio_utils.py

…

gemma3n_mm.py

Deprecate Transformers v4 support (#45161)

2026-06-11 11:04:01 +08:00

gemma3n.py

Remove KV cache scale boilerplate from model weight loading methods (#43167)

2026-06-05 05:19:04 -07:00

gemma4_mm.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

gemma4_mtp.py

[Model] Add Gemma4 Unified (encoder-free) support (#44429)

2026-06-03 12:01:39 -07:00

gemma4_unified.py

[Bugfix] Exclude vision embedder from quantization in Gemma4 Unified (#44571)

2026-06-04 20:47:38 -07:00

gemma4.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

gemma.py

Make Gemma and Gemma 2 accept inputs_embeds like Gemma 3 (#36787)

2026-03-11 18:12:43 +00:00

glm4_1v.py

[MM][Perf][CG] Support dual-path ViT full CUDA graph for DeepSeek-OCR (#43586)

2026-06-16 04:35:20 -07:00

glm4_moe_lite_mtp.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

glm4_moe_lite.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

glm4_moe_mtp.py

[ROCm][Perf] Add Fused Shared Expert (FSE) support for GLM-4.5/6/7 (#44313)

2026-06-28 00:04:08 -07:00

glm4_moe.py

[ROCm][Perf] Add Fused Shared Expert (FSE) support for GLM-4.5/6/7 (#44313)

2026-06-28 00:04:08 -07:00

glm4.py

Remove KV cache scale boilerplate from model weight loading methods (#43167)

2026-06-05 05:19:04 -07:00

glm4v.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

glm_ocr_mtp.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

glm_ocr.py

[bug-fix] GLM OCR Patch Merger context_dim (#37962)

2026-03-26 05:11:21 -07:00

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

[Audio] Bundle get_generation_prompt() params into SpeechToTextParams (#36268)

2026-04-22 12:24:18 +08:00

gpt2.py

[Bugfix][Model] Fix GPT2ForSequenceClassification sub-module prefix (#43579)

2026-05-25 22:43:19 -07:00

gpt_bigcode.py

[5/N][Attention] Finish eliminating vllm/attention folder (#32064)

2026-01-27 10:02:51 -05:00

gpt_j.py

Remove KV cache scale boilerplate from model weight loading methods (#43167)

2026-06-05 05:19:04 -07:00

gpt_neox.py

[5/N][Attention] Finish eliminating vllm/attention folder (#32064)

2026-01-27 10:02:51 -05:00

gpt_oss.py

[Bugfix]: Fix unquantized gpt-oss weight loading broken by FusedMoE r… (#45818)

2026-06-23 16:56:17 +00:00

granite4_vision.py

[Model] Fix missing maybe_prefix (#42280)

2026-05-11 09:04:06 +00:00

granite_speech_plus.py

Add model support for granite speech plus (#43519)

2026-06-04 14:47:48 +00:00

granite_speech.py

Add model support for granite speech plus (#43519)

2026-06-04 14:47:48 +00:00

granite.py

[Model] Remove InternLMForCausalLM registry alias (#45128)

2026-06-11 20:02:51 +00:00

granitemoe.py

Remove KV cache scale boilerplate from model weight loading methods (#43167)

2026-06-05 05:19:04 -07:00

granitemoehybrid.py

[MoE Refactor] FusedMoE/MoERunner inversion refactor (#41184)

2026-06-08 10:42:58 -04:00

granitemoeshared.py

[MoE Refactor] FusedMoE/MoERunner inversion refactor (#41184)

2026-06-08 10:42:58 -04:00

gritlm.py

[Perf] Remove redundant device copies for CPU-only pooling token IDs, 48.9% E2E throughput improvement (#38139)

2026-03-29 18:12:50 +00:00

h2ovl.py

[Model] Remove Mono-InternVL (InternLM2VEForCausalLM) (#45129)

2026-06-12 10:41:09 +00:00

hrm_text.py

[Model] Add HrmTextForCausalLM (Hierarchical Reasoning Model — Text) (#43098)

2026-06-16 22:41:41 +08:00

hunyuan_v1.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

hunyuan_vision.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

hy_v3_mtp.py

[Bugfix][MoE] Fix fused MoE expert mapping helper call sites (#44897)

2026-06-08 13:35:04 -07:00

hy_v3.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

hyperclovax_vision_v2.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

hyperclovax_vision.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

hyperclovax.py

Remove KV cache scale boilerplate from model weight loading methods (#43167)

2026-06-05 05:19:04 -07:00

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

[Perf][4/n] Eliminate various GPU<->CPU syncs (#42347)

2026-05-19 10:35:54 -04:00

interfaces_base.py

[Model] Deprecate the score task (this will not affect users). (#37537)

2026-03-20 08:07:56 +00:00

interfaces.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

intern_vit.py

[Model] Add torch.compile support for InternVL vision encoder (#38049)

2026-03-25 23:52:29 -07:00

internlm2.py

[Model] Use AutoWeightsLoader for InternLM2 (#38278)

2026-05-26 03:39:26 -07:00

interns1_pro.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

interns1_vit.py

[MM] Align the prefix of MMEncoderAttention with Attention (#33750)

2026-02-04 04:07:30 +00:00

interns1.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

interns2_preview.py

[Model] Support InternS2 Preview (#42705)

2026-05-14 21:30:26 -07:00

internvl.py

fix(quantization): Fix AWQ dequantize on Intel XPU and refactor AutoAWQ config (#42727)

2026-06-18 20:12:28 +08:00

iquest_loopcoder.py

Remove KV cache scale boilerplate from model weight loading methods (#43167)

2026-06-05 05:19:04 -07:00

isaac.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

jais2.py

[Migration] Migrate GGUF quantization support to plugin (#39612)

2026-06-12 12:02:21 -07:00

jamba.py

[MoE Refactor] Rename FusedMoE.make_expert_params_mapping to fused_moe_make_expert_params_mapping (#40671)

2026-04-23 11:22:34 -04:00

jina_vl.py

[Chore] Update type annotation of input_ids in model forward (#33063)

2026-01-26 06:02:10 -08:00

jina.py

Add Jina Embeddings v5 model support (fixes #38633) (#39575)

2026-04-16 06:37:10 +00:00

kanana_v.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

keye_vl1_5.py

[Model] Use mm_features for Keye-VL and Keye-1.5-VL M-RoPE (#39869)

2026-04-16 02:16:06 -07:00

keye.py

Remove KV cache scale boilerplate from model weight loading methods (#43167)

2026-06-05 05:19:04 -07:00

kimi_audio.py

fix: propagate revision/code_revision pins to all artifact boundaries (#42616)

2026-05-15 02:31:54 -07:00

kimi_k25_vit.py

Fix Kimi-K2.5 FlashInfer ViT metadata (#44493)

2026-06-04 08:14:35 +00:00

kimi_k25.py

Fix Kimi-K2.5 FlashInfer ViT metadata (#44493)

2026-06-04 08:14:35 +00:00

kimi_linear.py

[mamba] unify KDA conv states into one cache to match 2-state SSM layout (#44539)

2026-06-04 20:38:05 +02:00

kimi_vl.py

[MM][Perf][CG] Support ViT full CUDA graph for Kimi-VL (#41992)

2026-06-17 12:14:01 +00:00

laguna.py

Remove KV cache scale boilerplate from model weight loading methods (#43167)

2026-06-05 05:19:04 -07:00

lfm2_moe.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

lfm2_siglip2.py

[Model] Fix missing maybe_prefix (#42280)

2026-05-11 09:04:06 +00:00

lfm2_vl.py

[MM][Perf][CG] Support dual-path ViT full CUDA graph for DeepSeek-OCR (#43586)

2026-06-16 04:35:20 -07:00

lfm2.py

Fix Plamo 2/3 & LFM2 for Transformers v5 (#38090)

2026-03-25 12:29:49 +00:00

lightonocr.py

[Misc] Cleanup more configs and processors (#37560)

2026-03-19 15:45:23 +00:00

llama4_eagle.py

[SpecDecode] Reduce TP communication for large-vocab draft models speculative decoding (#39419)

2026-06-10 07:59:24 +00:00

llama4.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

llama_eagle3.py

Remove KV cache scale boilerplate from model weight loading methods (#43167)

2026-06-05 05:19:04 -07:00

llama_eagle.py

Remove KV cache scale boilerplate from model weight loading methods (#43167)

2026-06-05 05:19:04 -07:00

llama.py

[Migration] Migrate GGUF quantization support to plugin (#39612)

2026-06-12 12:02:21 -07:00

llava_next_video.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

llava_next.py

[Perf][4/n] Eliminate various GPU<->CPU syncs (#42347)

2026-05-19 10:35:54 -04:00

llava_onevision.py

[Perf][4/n] Eliminate various GPU<->CPU syncs (#42347)

2026-05-19 10:35:54 -04:00

llava.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

longcat_flash_mtp.py

[Chore] Update type annotation of input_ids in model forward (#33063)

2026-01-26 06:02:10 -08:00

longcat_flash.py

Refractor longcat loading to use AutoWeightsLoader (#41448)

2026-05-01 09:07:23 +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

[Chore] Update type annotation of input_ids in model forward (#33063)

2026-01-26 06:02:10 -08:00

medusa.py

…

mellum.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

midashenglm.py

Fix MiDashengLM TP>1 crash in audio encoder attention (#44408)

2026-06-09 09:29:14 -07:00

mimo_audio.py

[Model] Add MiMo-V2.5 support (#40967)

2026-04-27 13:26:51 +00:00

mimo_mtp.py

[Chore] Update type annotation of input_ids in model forward (#33063)

2026-01-26 06:02:10 -08:00

mimo_v2_mtp.py

[SpecDecoding] extend mtp support for mimo 2.5 (#41905)

2026-05-09 18:22:59 +00:00

mimo_v2_omni.py

[Model] Add MiMo-V2.5 support (#40967)

2026-04-27 13:26:51 +00:00

mimo_v2.py

[Models] Fix MiMo v2.x QKV TP sharding + FP4 support (#45200)

2026-06-15 15:13:34 +08:00

mimo.py

Remove KV cache scale boilerplate from model weight loading methods (#43167)

2026-06-05 05:19:04 -07:00

minicpm3.py

[5/N][Attention] Finish eliminating vllm/attention folder (#32064)

2026-01-27 10:02:51 -05:00

minicpm_eagle.py

…

minicpm.py

[MoE] Remove inplace fused experts mechanism (#43727)

2026-05-27 20:00:19 -07:00

minicpmo.py

[Bugfix] Fixes MiniCPM-O resampler device placement to avoid tensor device mismatch (#42332)

2026-06-17 08:35:27 +00:00

minicpmv4_6.py

minicpmv4_6: fix ImageSize (W,H) order for placeholder token calculation (#45244)

2026-06-11 13:43:56 +00:00

minicpmv.py

minicpmv4_6: fix ImageSize (W,H) order for placeholder token calculation (#45244)

2026-06-11 13:43:56 +00:00

minimax_m2.py

[PERF]MiniMax-M2 gate kernel (#38445)

2026-05-29 18:28:34 -07:00

mistral3.py

[Bugfix][Model] Fix Devstral Small 2 HF format weight loading (#39293)

2026-04-14 10:11:18 +00:00

mistral_eagle.py

[Model] Fix missing maybe_prefix (#42280)

2026-05-11 09:04:06 +00:00

mistral_large_3_eagle.py

[Bugfix] Initialize missing attributes in mistral eagle (#45217)

2026-06-12 08:21:01 +08:00

mistral_large_3.py

…

mistral.py

Remove unused logic in models/mistral.py (#33095)

2026-01-26 09:01:52 -08:00

mixtral.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

mllama4.py

[MM][Perf][CG] Support dual-path ViT full CUDA graph for DeepSeek-OCR (#43586)

2026-06-16 04:35:20 -07:00

mlp_speculator.py

[Refactor] Remove unused dead code (#38842)

2026-04-06 11:52:05 -04:00

modernbert.py

[ROCm][CI] Stabilize ModernBERT token-classification parity against Hugging Face (#44040)

2026-06-09 16:52:36 +01:00

module_mapping.py

[Misc] Remove unused ModelKeys (#32608)

2026-01-19 17:34:59 +00:00

molmo2.py

[Model] Fix missing maybe_prefix (#42280)

2026-05-11 09:04:06 +00:00

molmo.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

moondream3.py

[Model] Add Moondream3 model support(only query and caption skills) (#32325)

2026-05-01 10:06:48 +08:00

moonvit.py

[Core] Ensure memory is pinned prior to async h2d copy (#45424)

2026-06-20 20:02:24 -07:00

moss_audio.py

feat: support to OpenMOSS-Team (#44124)

2026-06-24 00:08:13 -07:00

mpt.py

[5/N][Attention] Finish eliminating vllm/attention folder (#32064)

2026-01-27 10:02:51 -05:00

musicflamingo.py

[ROCm][CI] Gate incompatible HF references on Transformers v5 (#41532)

2026-06-16 20:34:11 +08:00

nano_nemotron_vl.py

Avoid silent weights corruption when loading Nemotron Nano VL with reusable-buffer loaders like runai distributed streaming (#42244)

2026-05-11 12:03:14 +00:00

nemotron_h_mtp.py

[MoE Refactor] Rename FusedMoE.make_expert_params_mapping to fused_moe_make_expert_params_mapping (#40671)

2026-04-23 11:22:34 -04:00

nemotron_h.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

nemotron_nas.py

[Model] Remove InternLMForCausalLM registry alias (#45128)

2026-06-11 20:02:51 +00:00

nemotron_parse.py

[Bugfix] Fix LM detection for Nemotron Parse (#42641)

2026-05-14 23:42:10 +08:00

nemotron_vl.py

fix(quantization): Fix AWQ dequantize on Intel XPU and refactor AutoAWQ config (#42727)

2026-06-18 20:12:28 +08:00

nemotron.py

[Bugfix] Fix NemotronLayerNorm1P hardcoded cuda device type (#46495)

2026-06-24 13:21:02 +08:00

nvlm_d.py

[Model] Remove Mono-InternVL (InternLM2VEForCausalLM) (#45129)

2026-06-12 10:41:09 +00:00

olmo2.py

Speed up docs build (#44635)

2026-06-05 14:51:44 +00:00

olmo_hybrid.py

[Attention] Mamba attention module refactor (#41126)

2026-05-22 17:13:12 +08:00

olmo.py

Speed up docs build (#44635)

2026-06-05 14:51:44 +00:00

olmoe.py

[Migration] Migrate GGUF quantization support to plugin (#39612)

2026-06-12 12:02:21 -07:00

opencua.py

[Refactor] Define MM data parser in processing info instead of processor itself (#33260)

2026-01-29 13:55:17 +08:00

openpangu_mtp.py

[MoE Refactor] Rename FusedMoE.make_expert_params_mapping to fused_moe_make_expert_params_mapping (#40671)

2026-04-23 11:22:34 -04:00

openpangu_vl.py

[Quant] Consolidate GPTQ: rename gptq_marlin.py to auto_gptq.py (#38288)

2026-05-15 08:25:52 +08:00

openpangu.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

openvla.py

[Model] Openvla support (#42654)

2026-05-19 08:17:42 -07:00

opt.py

[5/N][Attention] Finish eliminating vllm/attention folder (#32064)

2026-01-27 10:02:51 -05:00

orion.py

[5/N][Attention] Finish eliminating vllm/attention folder (#32064)

2026-01-27 10:02:51 -05:00

ouro.py

Remove KV cache scale boilerplate from model weight loading methods (#43167)

2026-06-05 05:19:04 -07:00

ovis2_5.py

[Model] Fix missing maybe_prefix (#42280)

2026-05-11 09:04:06 +00:00

ovis.py

[Model] Fix missing maybe_prefix (#42280)

2026-05-11 09:04:06 +00:00

paddleocr_vl.py

Remove KV cache scale boilerplate from model weight loading methods (#43167)

2026-06-05 05:19:04 -07:00

paligemma.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

parakeet.py

Bugfix: Parakeet: .conv.pointwise/depthwise_conv1/2.bias weigths can exist even if convolution_bias=False (#40007)

2026-04-16 23:22:05 +00:00

param2moe.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

persimmon.py

[5/N][Attention] Finish eliminating vllm/attention folder (#32064)

2026-01-27 10:02:51 -05:00

phi3.py

…

phi3v.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

phi4mm_audio.py

[Perf][4/n] Eliminate various GPU<->CPU syncs (#42347)

2026-05-19 10:35:54 -04:00

phi4mm_utils.py

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

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

phi4mm.py

[ROCm][CI] Fix ROCm multimodal Qwen2.5-VL activation compile and Phi4MM ragged image mask handling (#43647)

2026-05-26 19:53:34 -07:00

phi4siglip.py

[Model] Add Phi4ForCausalLMV for microsoft/Phi-4-reasoning-vision-15B (#38306)

2026-04-02 21:14:57 -07:00

phi.py

[5/N][Attention] Finish eliminating vllm/attention folder (#32064)

2026-01-27 10:02:51 -05:00

phimoe.py

Remove KV cache scale boilerplate from model weight loading methods (#43167)

2026-06-05 05:19:04 -07:00

pixtral.py

[BUGFIX] Fix Pixtral consolidated format vision weight loading (#39916)

2026-04-19 22:25:03 -07:00

plamo2.py

[Misc] Replace mamba_type string literals with MambaAttentionBackendEnum (#41188)

2026-05-11 03:59:36 +00:00

plamo3.py

Fix Plamo 2/3 & LFM2 for Transformers v5 (#38090)

2026-03-25 12:29:49 +00:00

qianfan_ocr.py

[Model] support Qianfan-OCR model (#40136)

2026-05-05 10:51:25 +00:00

qwen2_5_omni_thinker.py

[Perf][4/n] Eliminate various GPU<->CPU syncs (#42347)

2026-05-19 10:35:54 -04:00

qwen2_5_vl.py

[Model Runner V2][MM] Support EVS (#46535)

2026-06-24 10:18:56 -04:00

qwen2_audio.py

[Perf][4/n] Eliminate various GPU<->CPU syncs (#42347)

2026-05-19 10:35:54 -04:00

qwen2_moe.py

[MoE Refactor] Rename FusedMoE.make_expert_params_mapping to fused_moe_make_expert_params_mapping (#40671)

2026-04-23 11:22:34 -04:00

qwen2_rm.py

[Chore] Update type annotation of input_ids in model forward (#33063)

2026-01-26 06:02:10 -08:00

qwen2_vl.py

[MM][Perf][CG] Support dual-path ViT full CUDA graph for DeepSeek-OCR (#43586)

2026-06-16 04:35:20 -07:00

qwen2.py

Remove KV cache scale boilerplate from model weight loading methods (#43167)

2026-06-05 05:19:04 -07:00

qwen3_5_mtp.py

[SpecDecode] Reduce TP communication for large-vocab draft models speculative decoding (#39419)

2026-06-10 07:59:24 +00:00

qwen3_5.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

qwen3_asr_forced_aligner.py

[Feature] Add Qwen3-ForcedAligner support via token classification pooling (#35367)

2026-03-29 00:27:52 +00:00

qwen3_asr_realtime.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

qwen3_asr.py

feat(qwen3-asr): support prompt parameter in v1/audio/transcriptions (#35415)

2026-06-10 19:54:59 +00:00

qwen3_dflash.py

[DFlash] Fuse precompute kv per-layer rmsnorms (#46761)

2026-06-25 19:32:07 -07:00

qwen3_eagle3.py

[Spec Decode] Add Qwen3 architecture support for EAGLE3 (#43132)

2026-06-21 13:55:26 -07:00

qwen3_moe.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

qwen3_next_mtp.py

[Bugfix] Fix NVFP4+MTP crash: force unquantized mtp.fc for Qwen3Next (#46316)

2026-06-25 09:05:04 -07:00

qwen3_next.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

qwen3_omni_moe_thinker.py

Deprecate Transformers v4 support (#45161)

2026-06-11 11:04:01 +08:00

qwen3_vl_moe.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

qwen3_vl.py

[Model Runner V2][MM] Support EVS (#46535)

2026-06-24 10:18:56 -04:00

qwen3.py

[SpecDecode] Reduce TP communication for large-vocab draft models speculative decoding (#39419)

2026-06-10 07:59:24 +00:00

radio.py

Fix Nano Nemotron VL regressions (#38655)

2026-04-03 15:22:06 +08:00

registry.py

[Model] Support Unlimited OCR (#46564)

2026-06-27 23:09:18 -07:00

rnj1.py

Remove KV cache scale boilerplate from model weight loading methods (#43167)

2026-06-05 05:19:04 -07:00

roberta.py

fix: propagate revision/code_revision pins to all artifact boundaries (#42616)

2026-05-15 02:31:54 -07:00

rvl.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

sarvam.py

[MoE Refactor] FusedMoE/MoERunner inversion refactor (#41184)

2026-06-08 10:42:58 -04:00

seed_oss.py

Remove KV cache scale boilerplate from model weight loading methods (#43167)

2026-06-05 05:19:04 -07:00

siglip2navit.py

[Model] Fix missing maybe_prefix (#42280)

2026-05-11 09:04:06 +00:00

siglip.py

[Migration] Migrate GGUF quantization support to plugin (#39612)

2026-06-12 12:02:21 -07:00

skyworkr1v.py

fix(quantization): Fix AWQ dequantize on Intel XPU and refactor AutoAWQ config (#42727)

2026-06-18 20:12:28 +08:00

smolvlm.py

[Bugfix] Standardize getting number of image patches/tokens (#34358)

2026-02-12 20:47:01 -08:00

solar.py

[Model] Remove InternLMForCausalLM registry alias (#45128)

2026-06-11 20:02:51 +00:00

stablelm.py

[5/N][Attention] Finish eliminating vllm/attention folder (#32064)

2026-01-27 10:02:51 -05:00

starcoder2.py

[5/N][Attention] Finish eliminating vllm/attention folder (#32064)

2026-01-27 10:02:51 -05:00

step1.py

[Refactor] Consolidate SupportsEagle (#36063)

2026-03-13 23:22:40 +00:00

step3_text.py

[Bugfix] Fix weight loading issues caused by #41184 (#45054)

2026-06-10 01:20:13 -04:00

step3_vl.py

[MM][Perf][CG] Support dual-path ViT full CUDA graph for DeepSeek-OCR (#43586)

2026-06-16 04:35:20 -07:00

step3p5_mtp.py

[Bugfix] LoRA: extend expert base_layer loading to Qwen3.5 and Step3.x (#37114)

2026-04-21 14:17:03 +00:00

step3p5.py

Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)

2026-06-27 00:16:05 +01:00

step3p7.py

[Model]Support Step-3.7-Flash (#43859)

2026-05-28 17:01:48 -07:00

step_vl.py

[MM][CG] Enable encoder Cudagraph for Step3VL (#42224)

2026-05-17 20:19:13 -07:00

tarsier.py

[Misc] Cleanup more configs and processors (#37560)

2026-03-19 15:45:23 +00:00

telechat2.py

…

teleflm.py

…

terratorch.py

[Misc] Reorganize inputs (#35182)

2026-03-25 10:22:54 -07:00

ultravox.py

Deprecate Transformers v4 support (#45161)

2026-06-11 11:04:01 +08:00

unlimited_ocr.py

[Model] Support Unlimited OCR (#46564)

2026-06-27 23:09:18 -07:00

utils.py

[CI Bug] Fix ValueError: There is no module or parameter named 'model.vision_tower.vision_model' (#45478)

2026-06-13 02:38:37 -07:00

vision.py

[Perf] FP8 FlashInfer Attn for ViT (#38065)

2026-04-27 13:44:15 +08:00

voxtral_realtime.py

Migrate Voxtral to mistral-common 1.11.5 audio API (#46705)

2026-06-26 11:06:31 -07:00

voxtral.py

Migrate Voxtral to mistral-common 1.11.5 audio API (#46705)

2026-06-26 11:06:31 -07:00

voyage.py

[Model] Use AutoWeightsLoader for Voyage (#42972)

2026-05-22 05:28:23 +00:00

whisper_causal.py

[Attention][V0 Deprecation] Deprecate accept output buffer (#39125)

2026-04-07 17:14:58 -04:00

whisper_utils.py

[Voxtral] Add new streaming arch (#32861)

2026-01-23 12:41:52 +01:00

whisper.py

[Audio] Bundle get_generation_prompt() params into SpeechToTextParams (#36268)

2026-04-22 12:24:18 +08:00

zamba2.py

[5/N][Attention] Finish eliminating vllm/attention folder (#32064)

2026-01-27 10:02:51 -05:00