exolabs/gpt-oss-120b-mxfp4-q8-dequant-bf16
Private dequantized BF16 export for vLLM validation.
- Internal model id:
gpt-oss-120b-mxfp4-q8
- MLX source:
mlx-community/gpt-oss-120b-MXFP4-Q8
- Original HF comparison model:
unsloth/gpt-oss-120b-BF16
- Export dtype:
bfloat16
- Text-only export:
False
This checkpoint is the result of dequantizing the quantized MLX checkpoint. It
is not the original upstream BF16 checkpoint.
