YihongJin
/

Qwen3-Omni-30B-A3B-Instruct-NVFP4-W4A16-max

8-bit precision

Model card Files Files and versions

Qwen3-Omni-30B-A3B-Instruct-NVFP4-W4A16-max

27.6 GB

Ctrl+K

Ctrl+K

1 contributor

History: 8 commits

YihongJin's picture

Remove --enforce-eager from serve command + cross-reference W4A4 siblings + vllm-omni#4025

1a0e6bd verified 3 months ago

.gitattributes

1.57 kB
Initial upload: NVFP4 W4A16 (max calibration, 1024 prompts) 3 months ago
README.md

4.9 kB
Remove --enforce-eager from serve command + cross-reference W4A4 siblings + vllm-omni#4025 3 months ago
chat_template.json

6.77 kB
add chat_template.json (missing from ModelOpt export) 3 months ago
config.json

14.7 kB
Initial upload: NVFP4 W4A16 (max calibration, 1024 prompts) 3 months ago
generation_config.json

182 Bytes
Initial upload: NVFP4 W4A16 (max calibration, 1024 prompts) 3 months ago
hf_quant_config.json

387 Bytes
Add *mlp.gate*, *lm_head* to exclude_modules (ModelOpt export dropped these) 3 months ago
model-00001-of-00003.safetensors

10 GB
xet

Initial upload: NVFP4 W4A16 (max calibration, 1024 prompts) 3 months ago
model-00002-of-00003.safetensors

9.84 GB
xet

Initial upload: NVFP4 W4A16 (max calibration, 1024 prompts) 3 months ago
model-00003-of-00003.safetensors

7.71 GB
xet

Initial upload: NVFP4 W4A16 (max calibration, 1024 prompts) 3 months ago
model.safetensors.index.json

6.65 MB
Initial upload: NVFP4 W4A16 (max calibration, 1024 prompts) 3 months ago
preprocessor_config.json

603 Bytes
add preprocessor_config.json (missing from ModelOpt export) 3 months ago
tokenizer.json

11.4 MB
xet

Initial upload: NVFP4 W4A16 (max calibration, 1024 prompts) 3 months ago
tokenizer_config.json

957 Bytes
Initial upload: NVFP4 W4A16 (max calibration, 1024 prompts) 3 months ago