Text Generation
MLX
Safetensors
qwen2
chat
conversational
Eval Results
8-bit precision