Where the 150 seconds actually go: profiling a local render loop
Splitting the 150 seconds into model load, sampling steps, and PNG encode with a stopwatch and cProfile, in a script short enough to paste My SD1.5 render loop runs in fp32 with model offload on a 4GB GTX 1650. A 512x768 image takes about 150 seconds, but that total does not identify the bottleneck…
Read the full story at DEV Community — AI ↗
Timeline · 1 report
- 2026-11-12 14:54 · DEV Community — AI
Where the 150 seconds actually go: profiling a local render loop