majentik/gpt-oss-120b-TurboQuant-MLX-4bit

Quantization vs openai/gpt-oss-120b. majentik/gpt-oss-120b-TurboQuant-MLX-4bit is a quantized version of the other (U32/U8).

1 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse