Binyao Jiang 24a8cee66d Fix GLM45v launch server cuda torch compile bug (#9554)

2025-08-25 13:46:28 -07:00

..

arcee.py

[Model] Add support for Arcee Foundational Model (#8154)

2025-07-30 10:45:25 -07:00

baichuan.py

Fix loading KV quantization scale; Enable modelopt kv cache (#4686)

2025-04-08 09:11:35 -07:00

bailing_moe.py

Support bailing moe (#8680)

2025-08-05 20:40:34 -07:00

bert.py

Feat/support rerank (#6058)

2025-06-16 10:50:01 -07:00

chatglm.py

Fix loading KV quantization scale; Enable modelopt kv cache (#4686)

2025-04-08 09:11:35 -07:00

clip.py

refactor: unify names of the feature field of MultimodalDataItem (#8075)

2025-07-16 17:52:38 -07:00

commandr.py

Revert "fix some typos" (#6244)

2025-05-12 12:53:26 -07:00

dbrx.py

[6/N] MoE Refactor: Cleanup MoE-related configs (#8849)

2025-08-14 21:14:53 -07:00

deepseek_janus_pro.py

refactor: unify names of the feature field of MultimodalDataItem (#8075)

2025-07-16 17:52:38 -07:00

deepseek_nextn.py

[PD] Support PD disaggregation with Prefill PP (#8846)

2025-08-16 18:31:31 -07:00

deepseek_v2.py

Overlapped weight offload (#8034)

2025-08-23 02:06:46 -07:00

deepseek_vl2.py

[Refactor] simplify multimodal data processing (#8107)

2025-07-20 21:43:09 -07:00

deepseek.py

[6/N] MoE Refactor: Cleanup MoE-related configs (#8849)

2025-08-14 21:14:53 -07:00

ernie4_eagle.py

Add ernie4.py for ERNIE-4.5 (#7657)

2025-08-08 00:55:48 -07:00

ernie4.py

[6/N] MoE Refactor: Cleanup MoE-related configs (#8849)

2025-08-14 21:14:53 -07:00

exaone.py

fix: EXAONE when using tie_word_embeddings (#5759)

2025-05-21 11:30:04 -07:00

gemma2_reward.py

[QUANT] Add GPTQModel Dynamic Quantization + lm_head Quantization (#3790)

2025-03-05 01:11:00 -08:00

gemma2.py

Fix incorrect default get_hidden_dim logic (#8987)

2025-08-09 00:25:38 -07:00

gemma3_causal.py

[Feature] Layer-wise Prefill (#7634)

2025-07-17 01:57:46 +08:00

gemma3_mm.py

refactor: unify names of the feature field of MultimodalDataItem (#8075)

2025-07-16 17:52:38 -07:00

gemma3n_audio.py

chore: remove vlm unnecessary import (#7541)

2025-06-26 01:38:15 -07:00

gemma3n_causal.py

Updates Gemma3n MLP layer to adapt latest transformers version (#7573)

2025-06-26 15:07:22 -07:00

gemma3n_mm.py

Improve LoRA Perf by Deprecating FlashInfer and Eliminating Redundant Tensor Ops (#8940)

2025-08-10 01:04:45 -07:00

gemma.py

[Feature] Layer-wise Prefill (#7634)

2025-07-17 01:57:46 +08:00

glm4_moe_nextn.py

[DP Attention] Refactor: adding some utility functions (#9136)

2025-08-13 21:08:06 -07:00

glm4_moe.py

Revert "[feature] Rework Ascend NPU graph support" (#9385)

2025-08-20 00:35:10 -07:00

glm4.py

Support glm4.1v and glm4.5v (#8798)

2025-08-09 00:59:13 -07:00

glm4v_moe.py

[GLM4.1V and GLM4.5V] Add vision transformer num_dummy_head support: max tp=4 -> max tp=8 (#9059)

2025-08-18 14:40:13 -07:00

glm4v.py

[GLM4.1V and GLM4.5V] Add vision transformer num_dummy_head support: max tp=4 -> max tp=8 (#9059)

2025-08-18 14:40:13 -07:00

gpt2.py

Fix loading KV quantization scale; Enable modelopt kv cache (#4686)

2025-04-08 09:11:35 -07:00

gpt_bigcode.py

Fix loading KV quantization scale; Enable modelopt kv cache (#4686)

2025-04-08 09:11:35 -07:00

gpt_oss.py

[fix] Fix mxfp4 triton MoE tp bug (#9473)

2025-08-23 01:48:40 -07:00

granite.py

Fix incorrect default get_hidden_dim logic (#8987)

2025-08-09 00:25:38 -07:00

granitemoe.py

[6/N] MoE Refactor: Cleanup MoE-related configs (#8849)

2025-08-14 21:14:53 -07:00

grok.py

fix(grok): remove duplicate replicate_lm_head configuration (#9549)

2025-08-23 19:49:24 -07:00

hunyuan.py

Support EPLB in FusedMoE (#8448)

2025-07-29 16:02:41 -07:00

idefics2.py

Fix incorrect LoRA weight loading for fused gate_up_proj (#6734)

2025-05-31 13:41:44 -07:00

internlm2_reward.py

[QUANT] Add GPTQModel Dynamic Quantization + lm_head Quantization (#3790)

2025-03-05 01:11:00 -08:00

internlm2.py

Support InternVL3 (#5350)

2025-05-01 22:38:59 -07:00

interns1.py

support for interns1-mini (#9299)

2025-08-18 17:56:04 -07:00

internvl.py

[GLM4.1V and GLM4.5V] Add vision transformer num_dummy_head support: max tp=4 -> max tp=8 (#9059)

2025-08-18 14:40:13 -07:00

kimi_vl_moonvit.py

Revert "fix some typos" (#6244)

2025-05-12 12:53:26 -07:00

kimi_vl.py

refactor: unify names of the feature field of MultimodalDataItem (#8075)

2025-07-16 17:52:38 -07:00

llama4.py

[6/N] MoE Refactor: Cleanup MoE-related configs (#8849)

2025-08-14 21:14:53 -07:00

llama_classification.py

[QUANT] Add GPTQModel Dynamic Quantization + lm_head Quantization (#3790)

2025-03-05 01:11:00 -08:00

llama_eagle3.py

support llama4 eagle3 (#6985)

2025-06-30 22:34:10 -07:00

llama_eagle.py

[PP] Add pipeline parallelism (#5724)

2025-04-30 18:18:07 -07:00

llama_embedding.py

[QUANT] Add GPTQModel Dynamic Quantization + lm_head Quantization (#3790)

2025-03-05 01:11:00 -08:00

llama_reward.py

[QUANT] Add GPTQModel Dynamic Quantization + lm_head Quantization (#3790)

2025-03-05 01:11:00 -08:00

llama.py

Feature: support qwen and llama4 reducescatter for dp attention padding (#9101)

2025-08-13 21:10:29 -07:00

llava.py

chore: bump transformer to 4.54.0 (#8416)

2025-07-27 21:27:25 -07:00

llavavid.py

refactor: unify names of the feature field of MultimodalDataItem (#8075)

2025-07-16 17:52:38 -07:00

mimo_mtp.py

Clean unused import for mimo mtp model (#7370)

2025-06-19 19:16:14 -07:00

mimo.py

Support XiaomiMiMo inference with mtp (#6059)

2025-05-22 14:14:49 -07:00

minicpm3.py

[6/N] MoE Refactor: Cleanup MoE-related configs (#8849)

2025-08-14 21:14:53 -07:00

minicpm.py

fix: remove redundant rotary embedding cache recomputation in MiniCPM (#8022)

2025-07-15 00:12:45 -07:00

minicpmo.py

chore: bump transformer to 4.54.0 (#8416)

2025-07-27 21:27:25 -07:00

minicpmv.py

refactor: unify names of the feature field of MultimodalDataItem (#8075)

2025-07-16 17:52:38 -07:00

mistral.py

refactor: unify names of the feature field of MultimodalDataItem (#8075)

2025-07-16 17:52:38 -07:00

mixtral_quant.py

[feature]Ascend quantization support (#7791)

2025-07-10 09:17:37 -07:00

mixtral.py

[6/N] MoE Refactor: Cleanup MoE-related configs (#8849)

2025-08-14 21:14:53 -07:00

mllama4.py

model: adapt mllama4 to VisionAttention (#8512)

2025-08-02 00:39:40 -07:00

mllama.py

Revert "[feature] Rework Ascend NPU graph support" (#9385)

2025-08-20 00:35:10 -07:00

nemotron_nas.py

from python.sglang.srt -> from sglang.srt (#9268)

2025-08-17 02:45:45 -07:00

olmo2.py

Fix loading KV quantization scale; Enable modelopt kv cache (#4686)

2025-04-08 09:11:35 -07:00

olmo.py

Fix loading KV quantization scale; Enable modelopt kv cache (#4686)

2025-04-08 09:11:35 -07:00

olmoe.py

[6/N] MoE Refactor: Cleanup MoE-related configs (#8849)

2025-08-14 21:14:53 -07:00

persimmon.py

Feat: Support for Persimmon Model (#7983)

2025-07-19 23:07:47 -07:00

phi3_small.py

Fix Phi3 serving which was broke by earlier change (#5991)

2025-05-03 00:28:47 -07:00

phi4mm_audio.py

Feat: Support audio in Phi4-mm model (#8048)

2025-07-18 21:03:53 -07:00

phi4mm_utils.py

Feat: Support audio in Phi4-mm model (#8048)

2025-07-18 21:03:53 -07:00

phi4mm.py

[CPU] Fix TP padding issue on Phi-4 (#8289)

2025-08-17 16:25:26 -07:00

phi.py

Support for Phi-1.5 & Phi-2 models (#7862)

2025-07-13 18:43:40 -07:00

phimoe.py

Support EPLB in FusedMoE (#8448)

2025-07-29 16:02:41 -07:00

pixtral.py

Refactor mm processors and Enable mixed modality processing (#7629)

2025-06-30 23:14:48 -07:00

qwen2_5_vl.py

Fix GLM45v launch server cuda torch compile bug (#9554)

2025-08-25 13:46:28 -07:00

qwen2_audio.py

Fix qwen2 audio not working bug (#8600)

2025-08-09 00:42:29 -07:00

qwen2_classification.py

[Model] Support Qwen2ForSequenceClassification (#4609)

2025-03-24 19:13:44 -07:00

qwen2_eagle.py

[Bugfix] pipeline parallelism and Eagle Qwen2 (#6910)

2025-06-07 01:58:50 -07:00

qwen2_moe.py

[6/N] MoE Refactor: Cleanup MoE-related configs (#8849)

2025-08-14 21:14:53 -07:00

qwen2_rm.py

[QUANT] Add GPTQModel Dynamic Quantization + lm_head Quantization (#3790)

2025-03-05 01:11:00 -08:00

qwen2_vl.py

refactor: unify names of the feature field of MultimodalDataItem (#8075)

2025-07-16 17:52:38 -07:00

qwen2.py

[DP Attention] Refactor: adding some utility functions (#9136)

2025-08-13 21:08:06 -07:00

qwen3_classification.py

Add support for Qwen3-seq-cls (#9357)

2025-08-20 16:51:56 -07:00

qwen3_moe.py

Revert "[feature] Rework Ascend NPU graph support" (#9385)

2025-08-20 00:35:10 -07:00

qwen3.py

Revert "[feature] Rework Ascend NPU graph support" (#9385)

2025-08-20 00:35:10 -07:00

qwen.py

[Feature] Layer-wise Prefill (#7634)

2025-07-17 01:57:46 +08:00

registry.py

Fix CI (#9012)

2025-08-09 13:33:42 -07:00

roberta.py

Feat/support rerank (#6058)

2025-06-16 10:50:01 -07:00

siglip.py

Implement Siglip Vision model, and support BNB quantization for gemma3-mm (#5339)

2025-05-20 23:53:46 -07:00

stablelm.py

Fix loading KV quantization scale; Enable modelopt kv cache (#4686)

2025-04-08 09:11:35 -07:00

step3_vl.py

[6/N] MoE Refactor: Cleanup MoE-related configs (#8849)

2025-08-14 21:14:53 -07:00

torch_native_llama.py

Fix incorrect default get_hidden_dim logic (#8987)

2025-08-09 00:25:38 -07:00

transformers.py

correct the tp_plan logic (#8850)

2025-08-07 16:53:34 -07:00

vila.py

refactor: unify names of the feature field of MultimodalDataItem (#8075)

2025-07-16 17:52:38 -07:00

xverse_moe.py

[6/N] MoE Refactor: Cleanup MoE-related configs (#8849)

2025-08-14 21:14:53 -07:00

xverse.py

Fix loading KV quantization scale; Enable modelopt kv cache (#4686)

2025-04-08 09:11:35 -07:00

yivl.py

[QUANT] Add GPTQModel Dynamic Quantization + lm_head Quantization (#3790)

2025-03-05 01:11:00 -08:00