Flux 2 Klein 9B with LoRA freezes at 70% sampling on a 3090 Ti, user seeks troubleshooting help
FearlessLeader17 · reddit · 2026-10-25
A user running Flux 2 Klein 9B (fp8) with qwen38b and flux2-vae on a 3090 Ti for consistent-character generation reports generation works fine without LoRAs (30s with one reference image, 2min with two), but any CivitAI-downloaded LoRA causes sampling to freeze at 70% with GPU usage spiking to 96-97%. They're unsure whether the randomly downloaded workflow or VRAM limits are to blame and ask for advice.
More from Multimodal
- Redditor shares EGO, a full-length AI-generated action thriller — primehouse_studio · 2026-10-25
- Krea CEO says open-source Krea 3 weights due about a week after release — East_Shoe8814 · 2026-10-25
- Can RTX 3060 run MiniMax H3 locally? Users weigh hardware requirements — Odd-Rip8379 · 2026-10-25
- Whispering Creek Tavern AI scene shows off character consistency — Some-Dark-5802 · 2026-10-25
- ElevenCreative offers $100k for the catchiest AI ad, sharing the full maker workflow — AIwithGhotai · 2026-10-25
- GPT-6 Astra generates a full real-estate video tour from a single Zillow listing — venturetwins · 2026-10-25