bearzi/gemma-4-E2B-it-oQ6
Text Generation • 1B • Updated • 54
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 13 days ago
MiMo-V2-Flash-JANG updated a collection 13 days ago
MiMo-V2-Flash-JANG updated a collection 13 days ago
MiMo-V2-Flash-JANGOrganizations
bearzi/gemma-4-E2B-it-oQ3
Text Generation • 0.8B • Updated • 59
bearzi/Trinity-Nano-Preview-oQ8
Text Generation • 2B • Updated • 46
bearzi/Trinity-Nano-Preview-oQ6
Text Generation • 1B • Updated • 18
bearzi/Trinity-Nano-Preview-oQ4
Text Generation • 1B • Updated • 14
bearzi/Qwen3-Coder-Next-oQ8
Text Generation • 22B • Updated • 971 • 1
bearzi/Qwen3-Coder-Next-oQ6
Text Generation • 18B • Updated • 833
bearzi/Qwen3-Coder-Next-oQ4
Text Generation • 13B • Updated • 848 • 2
bearzi/Qwen3-Coder-Next-oQ3
Text Generation • 10B • Updated • 702 • 1
bearzi/Qwen3-Coder-Next-oQ2
Text Generation • 8B • Updated • 569
bearzi/Qwen3.5-397B-A17B-oQ8
Text Generation • 108B • Updated • 51
bearzi/Qwen3.5-397B-A17B-oQ6
Text Generation • 87B • Updated • 40
bearzi/Qwen3.5-397B-A17B-oQ4
Text Generation • 64B • Updated • 46
bearzi/Qwen3.5-397B-A17B-oQ3
Text Generation • 51B • Updated • 33
bearzi/Qwen3.5-397B-A17B-oQ2
Text Generation • 40B • Updated • 57
bearzi/gemma-4-31b-it-oQ2
Text Generation • 4B • Updated • 45