Yohann Bearzi
bearzi
AI & ML interests
Uploading mixed-precision MLX quantizations of popular open-weight models, produced with oMLX’s oQ and JANG on apple silicon. Benchmarks and comparisons are welcome.
Recent Activity
updated a collection less than a minute ago
gemma-4-E4B-it-JANG updated a collection less than a minute ago
gemma-4-E4B-it-JANG updated a collection less than a minute ago
gemma-4-E4B-it-JANGOrganizations
models 132
bearzi/gemma-4-E4B-it-JANG_6M
Text Generation • 2B • Updated
bearzi/gemma-4-E4B-it-JANG_6K
Text Generation • 2B • Updated
bearzi/gemma-4-E4B-it-JANG_5K
Text Generation • 2B • Updated
bearzi/gemma-4-E4B-it-JANG_4S
Text Generation • 1B • Updated
bearzi/gemma-4-E4B-it-JANG_4M
Text Generation • 1B • Updated
bearzi/gemma-4-E4B-it-JANG_4L
Text Generation • 2B • Updated
bearzi/gemma-4-E4B-it-JANG_4K
Text Generation • 1B • Updated
bearzi/Trinity-Mini-oQ8
Text Generation • 7B • Updated
bearzi/Trinity-Mini-oQ6
Text Generation • 6B • Updated
bearzi/Trinity-Mini-oQ4
Text Generation • 4B • Updated