Qwen3.8-Flash-Next-EXL3-Framework-Strix-Halo-recipe
vcruz305/Qwen3.8-Flash-Next-EXL3-Framework-Strix-Halo-recipe
Python
Run Qwen3.8-Flash-Next EXL3 (turboderp pack) on one AMD Strix Halo / Framework Desktop (gfx1151) via the exllamav3-amd runtime fork. 41 tok/s mean, 47.5 peak, MTP self-speculation. Setup encodes the five ROCm/gfx1151 traps.