Krea 2 Local Generation Slower Than Expected
imataruto · reddit · 2026-07-19
The author tested the local generation speed of Krea 2 on an RTX 4090 and found it underperformed.
- Using krea2rawint8convrot + qwen3vl4bfp8scaled at 1MP resolution with 51 steps, the speed was about 4.19s/it, totaling 3 minutes 32 seconds per generation.
- With turbo enabled and 8 steps, it was about 2.2s/ts, total 16 seconds, which the author still finds slow.
- Peak VRAM usage was about 20GB, which doesn’t seem to be a bottleneck on a 24GB 4090.
- Trying Dynamic VRAM toggle and updating ComfyUI didn’t improve speed significantly, so the author suspects a configuration issue and asks if this is normal and what speeds others achieve with similar hardware.
More from Infra
- Legora runs 2,000 parallel agent evals on Modal Sandboxes for legal AI — blelbach · 2026-10-27
- Cloud release velocity scores: AWS 95, Azure 90, Google Cloud 82 — DavidLinthicum · 2026-10-27
- Selling a 5090 to scale local inference: one business weighs 5 hardware paths — Anarchaotic · 2026-10-27
- AWS tutorial: fleet-wide edge CV model lifecycle with SageMaker SDK v3, ~$0.07 per run — AWS ML Blog · 2026-10-27
- Qwen3.8-27B hits 40 tok/s on Jetson Thor with INT4 + MTP, accuracy beats FP16 — No-Dragonfly6246 · 2026-10-27
- Celesto launches CI runners nearly 90% cheaper than GitHub Actions, with agent sandboxes — aniketmaurya · 2026-10-27