Controlled Krea 2 character LoRA test: 0.5 MP is the sweet spot, 1.0 MP overbinds

Cold-Worldliness5392 · reddit · 2026-09-28

A controlled character-training experiment on Krea 2: the same 35-image Guy Fieri dataset trained as four variants (0.25/0.5/1.0 MP LoRA plus 0.5 MP LoKR with only LR bumped 5e-5→1e-4), all at 20 epochs with identical seeds/prompts/workflow on an RTX5090 via Fizgig on runpod.

Training times: 10 min (0.25 MP), 18 min (0.5 MP), 34 min (1.0 MP), 23 min (LoKR) — all far faster than older long-epoch workflows.

Key findings:

Recommendation: 0.5 MP + 20 epochs + rank 8 LoRA as production baseline; 0.5 MP LoKR worth further testing. A second-character comparison is underway.

Original post →

More from Multimodal

Multimodal channel →