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 about 6 hours ago
Qwen3.5-397B-A17B-JANG updated a collection about 6 hours ago
Qwen3.5-397B-A17B-JANG updated a collection about 6 hours ago
Qwen3.5-397B-A17B-JANGOrganizations
models 157
bearzi/Qwen3.5-397B-A17B-JANG_4L
58B • Updated
bearzi/Qwen3.5-397B-A17B-JANG_4M
58B • Updated
bearzi/Qwen3.5-397B-A17B-JANG_4S
57B • Updated
bearzi/Qwen3.5-397B-A17B-JANG_3L
53B • Updated
bearzi/Qwen3.5-397B-A17B-JANG_3M
53B • Updated
bearzi/Qwen3.5-397B-A17B-JANG_3S
53B • Updated
bearzi/Qwen3.5-397B-A17B-JANG_2L
54B • Updated • 1
bearzi/Qwen3.5-397B-A17B-JANG_2M
53B • Updated • 1
bearzi/Qwen3.5-397B-A17B-JANG_1L
54B • Updated
bearzi/Qwen3.5-397B-A17B-JANG_6M
83B • Updated