Reddit users say Krea 2 Turbo regains strong facial expressions with bypass LoRAs
YentaMagenta · reddit · 2026-07-21
- A Reddit post argues that vanilla Krea 2 Turbo is weaker on facial expressions because of censoring, but its native expression quality is still strong once bypass LoRAs are used.
- The author says you usually do not need highly specific anatomical prompts; concise, human-style expression descriptions work better.
- They recommend trying one bypass LoRA at a time at different strengths, and note that a simple prompt like “photo taken with a smartphone in 2021” can restore a more photorealistic look.
- A workflow link is provided, with base prompts promised in a comment.
More from Multimodal
- Fish Audio ships open-weight Drama 3 TTS; creator shows turning old posts into narrated videos — alex_verem · 2026-10-27
- Redditor crafts photorealistic 'morning bus ride' short film entirely in ComfyUI — no3us · 2026-10-27
- New blur/anti-blur slider LoRA for MOSS lets you dial background sharpness with weights — NoMouse9610 · 2026-10-27
- PixGR: Pixel-space generative decoding for fast feed-forward 3D reconstruction — kwangmoo_yi · 2026-10-27
- Krea-2 Turbo + ComfyUI for product ads: prompts and results shared — comfy-king-999 · 2026-10-27
- High-res tiled Marigold V2 depth maps arrive in ComfyUI with See Through model — AntonObukhov1 · 2026-10-27