bnellnm f9c069c85e Modularize fused experts and integrate PPLX kernels (#15956)

2025-05-14 13:11:54 -07:00

..

__init__.py

[V1]SupportsV0Only protocol for model definitions (#13959)

2025-02-27 20:02:15 -05:00

adapters.py

[Bugfix] Fix missing return value in load_weights method of adapters.py (#15542)

2025-03-31 06:56:42 -07:00

aimv2.py

[Model] Broadcast Ovis2 implementation to fit Ovis1.6 (#17861)

2025-05-11 17:56:30 -07:00

arctic.py

permute/unpermute kernel for moe optimization (#14568)

2025-05-02 11:31:55 -07:00

aria.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

aya_vision.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

baichuan.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

bamba.py

[Model] Mamba2 causal conv1d Refactor to Split Prefill and Decode Requests for Corresponding Kernels (#17146)

2025-05-06 17:59:30 -07:00

bart.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

bert_with_rope.py

[New Model]: support GTE NewModel (#17986)

2025-05-14 01:31:31 -07:00

bert.py

[New Model]: nomic-embed-text-v2-moe (#17785)

2025-05-11 00:59:43 -07:00

blip2.py

Add full API docs and improve the UX of navigating them (#17485)

2025-05-03 19:42:43 -07:00

blip.py

[SupportsQuant] Bert, Blip, Blip2, Bloom (#15573)

2025-04-03 08:23:19 -07:00

bloom.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

chameleon.py

[Fix] Support passing args to logger (#17425)

2025-04-30 08:06:58 -07:00

chatglm.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

clip.py

[VLM] Remove BaseProcessingInfo.get_mm_max_tokens_per_item (#16408)

2025-04-10 09:06:58 -07:00

commandr.py

[Model] Ignore rotary embed load for Cohere model (#17319)

2025-04-29 00:30:40 -07:00

constant_size_cache.py

[Model][MiniMaxText01] Support MiniMaxText01 model inference (#13454)

2025-04-01 16:23:55 -04:00

dbrx.py

Modularize fused experts and integrate PPLX kernels (#15956)

2025-05-14 13:11:54 -07:00

deepseek_mtp.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

deepseek_v2.py

Modularize fused experts and integrate PPLX kernels (#15956)

2025-05-14 13:11:54 -07:00

deepseek_vl2.py

[Optim] Compute multimodal hash only once per item (#17314)

2025-04-29 09:40:35 +08:00

deepseek.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

eagle.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

exaone.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

fairseq2_llama.py

[Misc] Add SPDX-License-Identifier headers to python source files (#12628)

2025-02-02 11:58:18 -08:00

falcon.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

florence2.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

fuyu.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

gemma2.py

[Bugfix] gemma[2,3] interleaved attention when sliding window is disabled (#17180)

2025-04-25 19:53:51 -07:00

gemma3_mm.py

[Bugfix] gemma[2,3] interleaved attention when sliding window is disabled (#17180)

2025-04-25 19:53:51 -07:00

gemma3.py

[Bugfix] gemma[2,3] interleaved attention when sliding window is disabled (#17180)

2025-04-25 19:53:51 -07:00

gemma.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

glm4.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

glm4v.py

[VLM] Avoid unnecessary dummy multimodal data during processing (#16416)

2025-04-10 19:32:14 +00:00

glm.py

[Bugfix] Fix GLM rotary_dim issue and support v1 (#16912)

2025-04-21 14:26:34 +00:00

gpt2.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

gpt_bigcode.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

gpt_j.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

gpt_neox.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

granite_speech.py

[Model] Add Granite Speech Support (#16246)

2025-04-28 10:05:00 +00:00

granite.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

granitemoe.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

granitemoehybrid.py

[Model] Mamba2 causal conv1d Refactor to Split Prefill and Decode Requests for Corresponding Kernels (#17146)

2025-05-06 17:59:30 -07:00

granitemoeshared.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

gritlm.py

[Model] GritLM supports other attention backends (#18109)

2025-05-14 03:33:19 -07:00

grok1.py

[Misc] Delete LoRA-related redundancy code (#17841)

2025-05-08 06:02:21 -07:00

h2ovl.py

[Optim] Compute multimodal hash only once per item (#17314)

2025-04-29 09:40:35 +08:00

idefics2_vision_model.py

[VLM] Clean up Phi-4-MM ViT implementation (#14812)

2025-03-15 18:53:52 -07:00

idefics3.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

interfaces_base.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

interfaces.py

Add full API docs and improve the UX of navigating them (#17485)

2025-05-03 19:42:43 -07:00

intern_vit.py

[Kernel] Use fused rmsnorm for some models like qwen3 series (#17735)

2025-05-06 23:10:02 -07:00

internlm2_ve.py

Remove unused kwargs from model definitions (#13555)

2025-02-24 17:13:52 -08:00

internlm2.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

internvl.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

jais.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

jamba.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

kimi_vl.py

[Bugfix] Add missing get_language_model to new MLLMs (#17300)

2025-04-28 04:09:57 -07:00

llama4.py

Modularize fused experts and integrate PPLX kernels (#15956)

2025-05-14 13:11:54 -07:00

llama_eagle3.py

[V1][Spec Decode] Share input embedding of target model with EAGLE draft model to free ~1GB for llama 3 model (#17326)

2025-05-14 12:31:46 -07:00

llama_eagle.py

[V1][Spec Decode] Share input embedding of target model with EAGLE draft model to free ~1GB for llama 3 model (#17326)

2025-05-14 12:31:46 -07:00

llama.py

[Model] GritLM supports other attention backends (#18109)

2025-05-14 03:33:19 -07:00

llava_next_video.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

llava_next.py

Add full API docs and improve the UX of navigating them (#17485)

2025-05-03 19:42:43 -07:00

llava_onevision.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

llava.py

Add full API docs and improve the UX of navigating them (#17485)

2025-05-03 19:42:43 -07:00

mamba2.py

[Model] Mamba2 causal conv1d Refactor to Split Prefill and Decode Requests for Corresponding Kernels (#17146)

2025-05-06 17:59:30 -07:00

mamba_cache.py

[Model][MiniMaxText01] Support MiniMaxText01 model inference (#13454)

2025-04-01 16:23:55 -04:00

mamba.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

medusa.py

[Misc] Add SPDX-License-Identifier headers to python source files (#12628)

2025-02-02 11:58:18 -08:00

mimo_mtp.py

[Model] Support MiMo-7B inference with MTP (#17433)

2025-05-12 23:25:33 +00:00

mimo.py

[Model] Support MiMo-7B inference with MTP (#17433)

2025-05-12 23:25:33 +00:00

minicpm3.py

Remove unused kwargs from model definitions (#13555)

2025-02-24 17:13:52 -08:00

minicpm.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

minicpmo.py

[Bugfix] Fix Minicpm-O-int4 GPTQ model inference (#17397)

2025-04-29 18:21:42 +00:00

minicpmv.py

[Bugfix] Fix Minicpm-O-int4 GPTQ model inference (#17397)

2025-04-29 18:21:42 +00:00

minimax_cache.py

[Model][MiniMaxText01] Support MiniMaxText01 model inference (#13454)

2025-04-01 16:23:55 -04:00

minimax_text_01.py

[Model] support MiniMax-VL-01 model (#16328)

2025-04-29 12:05:50 +08:00

minimax_vl_01.py

[Bugfix] Clean up MiniMax-VL and fix processing (#17354)

2025-04-29 20:42:16 +08:00

mistral3.py

Add full API docs and improve the UX of navigating them (#17485)

2025-05-03 19:42:43 -07:00

mixtral_quant.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

mixtral.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

mllama4.py

[Misc] Split model loader (#17712)

2025-05-07 12:42:26 +08:00

mllama.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

mlp_speculator.py

[Doc] Replace ibm-fms with ibm-ai-platform (#12709)

2025-02-04 07:05:04 +00:00

modernbert.py

[New Model]: support GTE NewModel (#17986)

2025-05-14 01:31:31 -07:00

module_mapping.py

[Misc] Add SPDX-License-Identifier headers to python source files (#12628)

2025-02-02 11:58:18 -08:00

molmo.py

[Kernel] Use fused rmsnorm for some models like qwen3 series (#17735)

2025-05-06 23:10:02 -07:00

moonvit.py

[Model][VLM] Add Kimi-VL model support (#16387)

2025-04-14 21:41:48 +00:00

mpt.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

nemotron_nas.py

[Misc] Delete LoRA-related redundancy code (#17841)

2025-05-08 06:02:21 -07:00

nemotron.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

nvlm_d.py

[VLM] Avoid unnecessary dummy multimodal data during processing (#16416)

2025-04-10 19:32:14 +00:00

olmo2.py

[Kernel] Use fused rmsnorm for some models like qwen3 series (#17735)

2025-05-06 23:10:02 -07:00

olmo.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

olmoe.py

[Fix] Support passing args to logger (#17425)

2025-04-30 08:06:58 -07:00

opt.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

orion.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

ovis.py

[Model] Broadcast Ovis2 implementation to fit Ovis1.6 (#17861)

2025-05-11 17:56:30 -07:00

paligemma.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

persimmon.py

[Minor][Models] Pass partial_rotary_factor parameter to rope (#17266)

2025-04-28 04:28:59 +00:00

phi3_small.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

phi3.py

[Misc] Add SPDX-License-Identifier headers to python source files (#12628)

2025-02-02 11:58:18 -08:00

phi3v.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

phi4mm_audio.py

[Model] Refactor Phi-4-multimodal to use merged processor and support V1 (#15477)

2025-04-19 02:26:11 -07:00

phi4mm_utils.py

Add full API docs and improve the UX of navigating them (#17485)

2025-05-03 19:42:43 -07:00

phi4mm.py

[Misc] Delete LoRA-related redundancy code (#17841)

2025-05-08 06:02:21 -07:00

phi.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

phimoe.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

pixtral.py

Add full API docs and improve the UX of navigating them (#17485)

2025-05-03 19:42:43 -07:00

plamo2.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

prithvi_geospatial_mae.py

[VLM] Avoid unnecessary dummy multimodal data during processing (#16416)

2025-04-10 19:32:14 +00:00

qwen2_5_omni_thinker.py

[Bugfix] use_fast failing to be propagated to Qwen2-VL image processor (#17838)

2025-05-08 05:39:21 -07:00

qwen2_5_vl.py

[Bugfix] use_fast failing to be propagated to Qwen2-VL image processor (#17838)

2025-05-08 05:39:21 -07:00

qwen2_audio.py

[Misc] Rename Audios -> Audio in Qwen2audio Processing (#17507)

2025-04-30 19:51:36 -07:00

qwen2_moe.py

Modularize fused experts and integrate PPLX kernels (#15956)

2025-05-14 13:11:54 -07:00

qwen2_rm.py

[V1]SupportsV0Only protocol for model definitions (#13959)

2025-02-27 20:02:15 -05:00

qwen2_vl.py

[Bugfix] use_fast failing to be propagated to Qwen2-VL image processor (#17838)

2025-05-08 05:39:21 -07:00

qwen2.py

[Model] GritLM supports other attention backends (#18109)

2025-05-14 03:33:19 -07:00

qwen3_moe.py

Modularize fused experts and integrate PPLX kernels (#15956)

2025-05-14 13:11:54 -07:00

qwen3.py

[Kernel] Use fused rmsnorm for some models like qwen3 series (#17735)

2025-05-06 23:10:02 -07:00

qwen_vl.py

Add full API docs and improve the UX of navigating them (#17485)

2025-05-03 19:42:43 -07:00

qwen.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

registry.py

[New Model]: support GTE NewModel (#17986)

2025-05-14 01:31:31 -07:00

roberta.py

[New Model]: nomic-embed-text-v2-moe (#17785)

2025-05-11 00:59:43 -07:00

siglip.py

[VLM] Remove BaseProcessingInfo.get_mm_max_tokens_per_item (#16408)

2025-04-10 09:06:58 -07:00

skyworkr1v.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

smolvlm.py

[Model] Add smolvlm support (#16017)

2025-04-08 19:12:17 -07:00

solar.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

stablelm.py

[Minor][Models] Pass partial_rotary_factor parameter to rope (#17266)

2025-04-28 04:28:59 +00:00

starcoder2.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

telechat2.py

Upstream Llama4 Support to Main (#16113)

2025-04-07 08:06:27 -07:00

teleflm.py

Upstream Llama4 Support to Main (#16113)

2025-04-07 08:06:27 -07:00

transformers.py

Transformers backend tweaks (#17365)

2025-04-29 09:08:03 -07:00

ultravox.py

[Misc] Split model loader (#17712)

2025-05-07 12:42:26 +08:00

utils.py

[Misc] Split model loader (#17712)

2025-05-07 12:42:26 +08:00

vision.py

Fix PixtralHF missing spatial_merge_size (#17571)

2025-05-01 22:14:09 -07:00

whisper.py

[Chore] Remove Sampler from Model Code (#17084)

2025-04-24 02:49:33 -07:00

zamba2.py

[Model] Mamba2 causal conv1d Refactor to Split Prefill and Decode Requests for Corresponding Kernels (#17146)

2025-05-06 17:59:30 -07:00