< cd ../gallery
[DevTools]

gemma4-12b-vllm-sm120

Reproducible recipe: serve abliterated Gemma-4-12B (gemma4_unified) at 50-118 tok/s on no-NVLink Blackwell (SM120) via vLLM nightly + ModelOpt FP8/NVFP4 + MTP spec-decode.
lna-lab
@lna-lab
maintainer
★ 15 stars
github
gemma4-12b-vllm-sm120 — preview
gemma4-12b-vllm-sm120 repository preview
# README
Reproducible recipe: serve abliterated Gemma-4-12B (gemma4_unified) at 50-118 tok/s on no-NVLink Blackwell (SM120) via vLLM nightly + ModelOpt FP8/NVFP4 + MTP spec-decode. Maintained by lna-lab on GitHub, where it has earned 15 stars from the community.
It's actively developed around abliterated, blackwell, fp8, and is a solid reference for anyone building with these tools.
# tags
# install
npm install gemma4-12b-vllm-sm120
languages
Python100%
last commit2 months ago
licenseApache-2.0
more vLLM repos
$ made-with-vllm
rssllmmadewithwhat

Ask MadeWithWhat

AI answers may contain mistakes — please double-check important details.