neural-nova
/

gemma-4-E4B-it-optimized

inference-optimization

Model card Files Files and versions

gemma-4-E4B-it-optimized / kernels /MLP /setup.py

Commit History

Add optimized CUDA kernels (RMSNorm, MLP) and patched Gemma4 modeling files

42e38d8
verified

Khanin Udomchoksakul commited on 4 days ago