xpuenabler/gpt-oss-15.5b-23E-SFT-v6-GpNd-separate-mode-decode-nf4
04
Initial commit: NF4 + AWQ quantization of mlp.experts (decode). Quantization parameters were computed on the prefill graph and reused on the decode graph.
initial commit
Initial commit: NF4 + AWQ quantization of mlp.experts (decode). Quantization parameters were computed on the prefill graph and reused on the decode graph.
initial commit