batiai/DeepSeek-V4-Flash-GGUF
Text Generation • 284B • Updated • 1.2k • 3
🚀 Frontier MoE — 128B–1T
Largest open-weight LLMs, BatiAI-quantized. Mac-runnable from M4 Max 128GB to Mac Studio M3 Ultra 512GB.
Note 284B-A13B MoE • CSA+HCA hybrid attention • SWE-Bench Pro top tier • via batiai/bati.cpp
-
batiai/Kimi-K2.6-GGUF
Text Generation • 1T • Updated • 6.11k -
batiai/MiniMax-M2.7-GGUF
Text Generation • 229B • Updated • 1.31k
batiai/Mistral-Medium-3.5-128B-GGUF
Text Generation • 125B • Updated • 659Note 128B Dense • SWE-Bench Verified 77.6% • Modified MIT • measured 6.8 t/s on M4 Max IQ3