Luciano Martins 1317028aa8

[Model] Gemma3: Fix GGUF loading and quantization (#26189) ...

Signed-off-by: Luciano Martins <lucianommartins@users.noreply.github.com>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Luciano Martins <lucianommartins@users.noreply.github.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>

2025-10-09 07:00:53 +00:00

..

__init__.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

adapters.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

aimv2.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

apertus.py

[Models] Improve iteration over layers (#26425)

2025-10-08 20:48:33 +00:00

arcee.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

arctic.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

aria.py

Remove all references to yapf as it's no longer used (#26251)

2025-10-05 09:18:11 -07:00

aya_vision.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

baichuan.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

bailing_moe.py

Fix per file ruff ignores related to line length (#26262)

2025-10-06 05:12:40 +00:00

bamba.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

bert_with_rope.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

bert.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

blip2.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

blip.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

bloom.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

chameleon.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

chatglm.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

clip.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

cohere2_vision.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

commandr.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

config.py

[Hybrid]: Decouple Kernel Block Size from KV Page Size (#24486)

2025-10-08 23:43:39 -07:00

dbrx.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

deepseek_eagle.py

[Bugfix] Allow skipping MoE in NVFP4 (fix for MTP) (#25987)

2025-10-06 16:16:30 -04:00

deepseek_mtp.py

[Bugfix] Allow skipping MoE in NVFP4 (fix for MTP) (#25987)

2025-10-06 16:16:30 -04:00

deepseek_v2.py

Separate MLAAttention class from Attention (#25103)

2025-10-08 17:11:11 -07:00

deepseek_vl2.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

deepseek.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

dots1.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

dots_ocr.py

Fix DotsOCR tensor type (#26281)

2025-10-06 12:23:43 +00:00

ernie45_moe.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

ernie45_vl_moe.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

ernie45_vl.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

ernie45.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

ernie_mtp.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

exaone4.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

exaone.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

fairseq2_llama.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

falcon_h1.py

[Models] Improve iteration over layers (#26425)

2025-10-08 20:48:33 +00:00

falcon.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

fuyu.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

gemma2.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

gemma3_mm.py

Remove all references to yapf as it's no longer used (#26251)

2025-10-05 09:18:11 -07:00

gemma3.py

[Model] Gemma3: Fix GGUF loading and quantization (#26189)

2025-10-09 07:00:53 +00:00

gemma3n_mm.py

[Misc] Clean up unnecessary E501 ignore (#26274)

2025-10-06 07:29:18 +00:00

gemma3n.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

gemma.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

glm4_1v.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

glm4_moe_mtp.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

glm4_moe.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

glm4.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

glm4v.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

glm.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

gpt2.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

gpt_bigcode.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

gpt_j.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

gpt_neox.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

gpt_oss.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

granite_speech.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

granite.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

granitemoe.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

granitemoehybrid.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

granitemoeshared.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

gritlm.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

grok1.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

h2ovl.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

hunyuan_v1.py

[Models] Improve iteration over layers (#26425)

2025-10-08 20:48:33 +00:00

hyperclovax_vision.py

Fix per file ruff ignores related to line length (#26262)

2025-10-06 05:12:40 +00:00

idefics2_vision_model.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

idefics3.py

Remove all references to yapf as it's no longer used (#26251)

2025-10-05 09:18:11 -07:00

interfaces_base.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

interfaces.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

intern_vit.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

internlm2_ve.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

internlm2.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

interns1_vit.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

interns1.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

internvl.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

jais.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

jamba.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

jina_vl.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

keye_vl1_5.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

keye.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

kimi_vl.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

lfm2_moe.py

[Models] Improve iteration over layers (#26425)

2025-10-08 20:48:33 +00:00

lfm2.py

[Model] Lfm2Moe (#26344)

2025-10-07 16:03:05 +00:00

llama4_eagle.py

Fix per file ruff ignores related to line length (#26262)

2025-10-06 05:12:40 +00:00

llama4.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

llama_eagle3.py

Support llama3 eagle3 head with llama4 verifier (#25961)

2025-10-06 13:56:08 -04:00

llama_eagle.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

llama.py

Support llama3 eagle3 head with llama4 verifier (#25961)

2025-10-06 13:56:08 -04:00

llava_next_video.py

[Model] Use merge_by_field_config for MM models (Llava family) (#26280)

2025-10-06 09:45:26 +00:00

llava_next.py

[Model] Use merge_by_field_config for MM models (Llava family) (#26280)

2025-10-06 09:45:26 +00:00

llava_onevision.py

[Model] Use merge_by_field_config for MM models (Llava family) (#26280)

2025-10-06 09:45:26 +00:00

llava.py

[Model] Use merge_by_field_config for MM models (Llava family) (#26280)

2025-10-06 09:45:26 +00:00

longcat_flash_mtp.py

Fix per file ruff ignores related to line length (#26262)

2025-10-06 05:12:40 +00:00

longcat_flash.py

[Models] Improve iteration over layers (#26425)

2025-10-08 20:48:33 +00:00

mamba2.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

mamba.py

[Models] Improve iteration over layers (#26425)

2025-10-08 20:48:33 +00:00

medusa.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

midashenglm.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

mimo_mtp.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

mimo.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

minicpm3.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

minicpm_eagle.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

minicpm.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

minicpmo.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

minicpmv.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

minimax_text_01.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

minimax_vl_01.py

[Model] Use merge_by_field_config for MM models (Llava family) (#26280)

2025-10-06 09:45:26 +00:00

mistral3.py

[Model] Use merge_by_field_config for MM models (Llava family) (#26280)

2025-10-06 09:45:26 +00:00

mixtral.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

mllama4.py

Support llama3 eagle3 head with llama4 verifier (#25961)

2025-10-06 13:56:08 -04:00

mlp_speculator.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

modernbert.py

[Model] Add support for ModernBertForTokenClassification (#26340)

2025-10-07 14:29:19 +00:00

module_mapping.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

molmo.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

moonvit.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

mpt.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

nano_nemotron_vl.py

[Model] Allow passing custom number of max tiles to Nano 2 VL (#26403)

2025-10-08 11:19:39 +00:00

nemotron_h.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

nemotron_nas.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

nemotron_vl.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

nemotron.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

nvlm_d.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

olmo2.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

olmo.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

olmoe.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

opt.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

orion.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

ovis2_5.py

[Model] Use merge_by_field_config for MM models (Ovis family) (#26308)

2025-10-07 12:54:22 +00:00

ovis.py

[Model] Use merge_by_field_config for MM models (Ovis family) (#26308)

2025-10-07 12:54:22 +00:00

paligemma.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

persimmon.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

phi3.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

phi3v.py

Remove all references to yapf as it's no longer used (#26251)

2025-10-05 09:18:11 -07:00

phi4_multimodal.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

phi4mm_audio.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

phi4mm_utils.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

phi4mm.py

Fix per file ruff ignores related to line length (#26262)

2025-10-06 05:12:40 +00:00

phi.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

phimoe.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

pixtral.py

[Model] Use merge_by_field_config for MM models (Llava family) (#26280)

2025-10-06 09:45:26 +00:00

plamo2.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

qwen2_5_omni_thinker.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

qwen2_5_vl.py

[Model] EVS support for nano_nemotron_vl (#26269)

2025-10-07 00:23:37 +08:00

qwen2_audio.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

qwen2_moe.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

qwen2_rm.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

qwen2_vl.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

qwen2.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

qwen3_moe.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

qwen3_next_mtp.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

qwen3_next.py

Fix per file ruff ignores related to line length (#26262)

2025-10-06 05:12:40 +00:00

qwen3_vl_moe.py

[Models] Improve iteration over layers (#26425)

2025-10-08 20:48:33 +00:00

qwen3_vl.py

[Models] Improve iteration over layers (#26425)

2025-10-08 20:48:33 +00:00

qwen3.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

qwen_vl.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

qwen.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

radio.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

registry.py

[Misc] add usedforsecurity=False in md5 hash call (#26357)

2025-10-08 10:18:32 +00:00

roberta.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

rvl.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

seed_oss.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

siglip2navit.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

siglip.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

skyworkr1v.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

smolvlm.py

Remove all references to yapf as it's no longer used (#26251)

2025-10-05 09:18:11 -07:00

solar.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

stablelm.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

starcoder2.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

step3_text.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

step3_vl.py

[Model] Define merge_by_field_config MM interface (R-T) (#26260)

2025-10-07 16:10:55 +08:00

swin.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

tarsier.py

[Model] Define merge_by_field_config MM interface (R-T) (#26260)

2025-10-07 16:10:55 +08:00

telechat2.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

teleflm.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

terratorch.py

[Model] Define merge_by_field_config MM interface (R-T) (#26260)

2025-10-07 16:10:55 +08:00

transformers_moe.py

[Bugfix] Fix gemma3 with transformers backend (#23178)

2025-10-06 18:42:32 +00:00

transformers_pooling.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

transformers.py

[Bugfix] Fix gemma3 with transformers backend (#23178)

2025-10-06 18:42:32 +00:00

ultravox.py

[Model] Define merge_by_field_config MM interface (U-Z) (#26261)

2025-10-07 06:45:49 +00:00

utils.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

vision.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

voxtral.py

[Model] Define merge_by_field_config MM interface (U-Z) (#26261)

2025-10-07 06:45:49 +00:00

whisper.py

[Model] Define merge_by_field_config MM interface (U-Z) (#26261)

2025-10-07 06:45:49 +00:00

zamba2.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00