local intelligence

can it run

Can the Radeon 660M run SDXL SDXL base?

FULL

8.9 GB needed of 13.5 GB usable — headroom for context

~13.5GB usable · RDNA2 (iGPU) · model file 6.9GB · SDXL base

~2.8–8.9 tok/s

computed roofline: 76.8 GB/s × 0.25–0.80 efficiency window / 6.9GB (Q4_K_M) — bands, never points.

Run SDXL SDXL base on instead

Tesla P100 (16GB, used) — cheapest catalog machine that runs SDXL SDXL base fully ($135)

every model the Radeon 660M can run — full list

explore the full catalog84 machines indexed · live prices · what each one can run
2026-09-17 · ← all models on the Radeon 660M · fit = weights + 4k KV vs usable memory