Hugging Face's logo

majentik
/
Qwen3.5-27B-RotorQuant-MLX-4bit

MLX
Safetensors
qwen3_5
mlx-lm
rotorquant
kv-cache-quantization
qwen3.5
thinking-model
4bit
quantized
4-bit precision
Model card Files
xet
 Community
Qwen3.5-27B-RotorQuant-MLX-4bit / tokenizer.json

Commit History

Add MLX 4-bit quantized Qwen3.5-27B with RotorQuant KV cache
aca626f
verified

majentik commited on