cloudnathan5
/

Muse-Glimmer-30B-NVFP4

Image-Text-to-Text

compressed-tensors

8-bit precision

Model card Files Files and versions

Muse-Glimmer-30B-NVFP4

23.4 GB

Ctrl+K

Ctrl+K

1 contributor

History: 14 commits

cloudnathan5's picture

Remove stale fork patch: superseded by xianbaoqian/vllm#1 (fix-spec-decode), which the card now points to

2c6a808 verified about 8 hours ago

.gitattributes

1.57 kB
Upload folder using huggingface_hub 1 day ago
LICENSE

11.4 kB
Upload folder using huggingface_hub 1 day ago
README.md

14.6 kB
Fix install step: plain editable install, not --no-deps (matches the validated setup) about 9 hours ago
USAGE_POLICY.md

5.23 kB
Upload folder using huggingface_hub 1 day ago
chat_template.jinja

7.17 kB
Upload folder using huggingface_hub 1 day ago
config.json

6.3 kB
Make ignore list portable: name the unquantized vision stack explicitly (SGLang compatibility) about 17 hours ago
generation_config.json

185 Bytes
Upload folder using huggingface_hub 1 day ago
model-00001-of-00002.safetensors

20 GB
xet

GPTQ-rounded NVFP4A16: PPL 5.5631 (was 5.5965 RTN), 139.3 tok/s with DFlash about 23 hours ago
model-00002-of-00002.safetensors

3.38 GB
xet

GPTQ-rounded NVFP4A16: PPL 5.5631 (was 5.5965 RTN), 139.3 tok/s with DFlash about 23 hours ago
model.safetensors.index.json

224 kB
Replace W4A4 with NVFP4A16: faster (74.7 vs 54.3 tok/s) and better PPL (5.5965 vs 5.7660) 1 day ago
processor_config.json

1.08 kB
Upload folder using huggingface_hub 1 day ago
recipe.yaml

579 Bytes
GPTQ-rounded NVFP4A16: PPL 5.5631 (was 5.5965 RTN), 139.3 tok/s with DFlash about 23 hours ago
tokenizer.json

28.1 MB
xet

Replace W4A4 with NVFP4A16: faster (74.7 vs 54.3 tok/s) and better PPL (5.5965 vs 5.7660) 1 day ago
tokenizer_config.json

80 kB
Upload folder using huggingface_hub 1 day ago