Lukas Geiger 07cadab27a

[Model][Qwen3VL] Cache positional embedding indices (#28475) ...

Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>
Co-authored-by: Roger Wang <hey@rogerw.io>

2025-11-15 19:03:09 +00:00

..

transformers

[Docs] Update the name of Transformers backend -> Transformers modeling backend (#28725)

2025-11-14 16:34:14 +00:00

__init__.py

…

adapters.py

[Docs] Update the name of Transformers backend -> Transformers modeling backend (#28725)

2025-11-14 16:34:14 +00:00

aimv2.py

…

apertus.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

arcee.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

arctic.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

aria.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

aya_vision.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

baichuan.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

bailing_moe.py

[Model] Fix lmhead init bug of bailing_moe (#28777)

2025-11-15 05:44:12 -08:00

bamba.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

bee.py

…

bert_with_rope.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

bert.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

blip2.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

blip.py

…

bloom.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

chameleon.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

chatglm.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

clip.py

[Model][MM] Extract conv layer as CustomOp (#28455)

2025-11-14 19:16:13 +08:00

cohere2_vision.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

commandr.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

config.py

[Performance][gpt-oss] Revert gpt-oss max cudagraph size to 1024 (#28345)

2025-11-08 15:50:10 -07:00

dbrx.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

deepencoder.py

…

deepseek_eagle.py

Add support for Eagle with separate lm-head and embed_tokens layers (#28549)

2025-11-15 06:12:02 -08:00

deepseek_mtp.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

deepseek_ocr.py

[BugFix] DeepSeek-OCR: apply NoRepeatNGramLogitsProcessor to greedy path (#28617)

2025-11-13 15:24:35 +00:00

deepseek_v2.py

Add support for Eagle with separate lm-head and embed_tokens layers (#28549)

2025-11-15 06:12:02 -08:00

deepseek_vl2.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

dots1.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

dots_ocr.py

[Bugfix] fix dots.ocr pp support (#28705)

2025-11-14 09:01:26 +00:00

ernie45_moe.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

ernie45_vl_moe.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

ernie45_vl.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

ernie45.py

[Model] Add PaddleOCR-VL Model Support (#27758)

2025-11-03 19:04:22 +08:00

ernie_mtp.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

exaone4.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

exaone.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

fairseq2_llama.py

…

falcon_h1.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

falcon.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

flex_olmo.py

…

fuyu.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

gemma2.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

gemma3_mm.py

Allow Gemma3 to take image embeddings (#28483)

2025-11-15 04:18:08 -08:00

gemma3.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

gemma3n_mm.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

gemma3n.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

gemma.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

glm4_1v.py

[Model][MM] Extract conv layer as CustomOp (#28455)

2025-11-14 19:16:13 +08:00

glm4_moe_mtp.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

glm4_moe.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

glm4.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

glm4v.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

glm.py

…

gpt2.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

gpt_bigcode.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

gpt_j.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

gpt_neox.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

gpt_oss.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

granite_speech.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

granite.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

granitemoe.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

granitemoehybrid.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

granitemoeshared.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

gritlm.py

…

grok1.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

h2ovl.py

…

hunyuan_v1.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

hyperclovax_vision.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

idefics2_vision_model.py

…

idefics3.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

interfaces_base.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

interfaces.py

Add support for Eagle with separate lm-head and embed_tokens layers (#28549)

2025-11-15 06:12:02 -08:00

intern_vit.py

…

internlm2_ve.py

…

internlm2.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

interns1_vit.py

…

interns1.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

internvl.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

jais.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

jamba.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

jina_vl.py

…

keye_vl1_5.py

[Model] Pass mm_features directly into get_mrope_input_positions (#28399)

2025-11-11 21:14:48 +08:00

keye.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

kimi_linear.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

kimi_vl.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

lfm2_moe.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

lfm2.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

lightonocr.py

…

llama4_eagle.py

Add support for Eagle with separate lm-head and embed_tokens layers (#28549)

2025-11-15 06:12:02 -08:00

llama4.py

[BugFix] Support EP/DP + EPLB with MTP (#25311)

2025-11-05 15:22:17 +00:00

llama_eagle3.py

Add support for Eagle with separate lm-head and embed_tokens layers (#28549)

2025-11-15 06:12:02 -08:00

llama_eagle.py

Add support for Eagle with separate lm-head and embed_tokens layers (#28549)

2025-11-15 06:12:02 -08:00

llama.py

Add support for Eagle with separate lm-head and embed_tokens layers (#28549)

2025-11-15 06:12:02 -08:00

llava_next_video.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

llava_next.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

llava_onevision.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

llava.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

longcat_flash_mtp.py

…

longcat_flash.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

mamba2.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

mamba.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

medusa.py

[LoRA][1/N]Remove LoRA extra vocab (#28382)

2025-11-11 11:06:21 -08:00

midashenglm.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

mimo_mtp.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

mimo.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

minicpm3.py

[Misc][Model][Refactor] Pass the prefix into Linear layers (#28259)

2025-11-07 19:38:38 +08:00

minicpm_eagle.py

Add support for Eagle with separate lm-head and embed_tokens layers (#28549)

2025-11-15 06:12:02 -08:00

minicpm.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

minicpmo.py

…

minicpmv.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

minimax_m2.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

minimax_text_01.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

minimax_vl_01.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

mistral3.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

mixtral.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

mllama4.py

LLaMA4 LoRA Adapter Enablement (#28602)

2025-11-14 13:27:56 -05:00

mlp_speculator.py

[LoRA][1/N]Remove LoRA extra vocab (#28382)

2025-11-11 11:06:21 -08:00

modernbert.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

module_mapping.py

…

molmo.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

moonvit.py

[CI/Build] Bump transformers version (#27528)

2025-10-31 22:11:07 -07:00

mpt.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

nano_nemotron_vl.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

nemotron_h.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

nemotron_nas.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

nemotron_vl.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

nemotron.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

nvlm_d.py

…

olmo2.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

olmo.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

olmoe.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

openpangu_mtp.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

openpangu.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

opt.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

orion.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

ouro.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

ovis2_5.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

ovis.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

paddleocr_vl.py

[Fix] improve aspect ratio in dummy image generation and add common VLM tests for PaddleOCR-VL (#28711)

2025-11-14 16:07:20 +00:00

paligemma.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

persimmon.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

phi3.py

…

phi3v.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

phi4_multimodal.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

phi4mm_audio.py

…

phi4mm_utils.py

…

phi4mm.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

phi.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

phimoe.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

pixtral.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

plamo2.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

qwen2_5_omni_thinker.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

qwen2_5_vl.py

[Model][MM] Extract conv layer as CustomOp (#28455)

2025-11-14 19:16:13 +08:00

qwen2_audio.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

qwen2_moe.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

qwen2_rm.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

qwen2_vl.py

[Model][MM] Extract conv layer as CustomOp (#28455)

2025-11-14 19:16:13 +08:00

qwen2.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

qwen3_moe.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

qwen3_next_mtp.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

qwen3_next.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

qwen3_omni_moe_thinker.py

[Model][MM] Extract conv layer as CustomOp (#28455)

2025-11-14 19:16:13 +08:00

qwen3_vl_moe.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

qwen3_vl.py

[Model][Qwen3VL] Cache positional embedding indices (#28475)

2025-11-15 19:03:09 +00:00

qwen3.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

qwen_vl.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

qwen.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

radio.py

…

registry.py

[Model] Consolidate Deepseek-MoE implementation with DeepSeek-v2 (#28101)

2025-11-08 05:01:27 +00:00

roberta.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

rvl.py

…

seed_oss.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

siglip2navit.py

[BugFix] Fix Siglip2Attention on XPU (#28448)

2025-11-11 18:18:02 +00:00

siglip.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

skyworkr1v.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

smolvlm.py

…

solar.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

stablelm.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

starcoder2.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

step3_text.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

step3_vl.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

swin.py

…

tarsier.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

telechat2.py

…

teleflm.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

terratorch.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

ultravox.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

utils.py

Add support for Eagle with separate lm-head and embed_tokens layers (#28549)

2025-11-15 06:12:02 -08:00

vision.py

[Model][MM] Extract conv layer as CustomOp (#28455)

2025-11-14 19:16:13 +08:00

voxtral.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

whisper.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00

zamba2.py

Rename clashing method names for vLLM model protocol (#27583)

2025-11-12 19:14:33 -08:00