Hugging Face's logo

majentik
/
Mistral-Small-4-119B-RotorQuant-MLX-4bit

Text Generation
MLX
Safetensors
mistral3
rotorquant
kv-cache-quantization
mistral
Mixture of Experts
sparse-moe
multimodal
quantized
4-bit precision
apple-silicon
256k-context
thinking
conversational
Model card Files
xet
 Community
Mistral-Small-4-119B-RotorQuant-MLX-4bit / params.json

Commit History

Add MLX 4-bit quantized Mistral Small 4 with KV cache compression
9968358
verified

majentik commited on