Kimi-K2.6 - a baa-ai Collection

baa-ai 's Collections

LTX-2.3 RAM Quantized (MLX)

Qwen3.6-35B-A3B

DeepSeek-V3.2 (MLX)

MiniMax-M2.7 (MLX)

Nemotron 3 Super

Qwen3.5-35B-A3B

Qwen3.5-122B-A10B

Qwen3.5-397B-A17B

Kimi-K2.6

updated Apr 21

Mixed-precision GGUF quantizations of moonshotai/Kimi-K2.6 from the RAM pipeline (per-tensor bit allocation via sensitivity probing).