<h3 id="klein-9b">Klein 9B</h3>Showcase images :<ul><li>Distilled model</li><li>768x1152</li><li>Euler / normal / 4 steps / CFG 1</li><li>Lora strength of 1</li></ul><h3 id="zib">ZIB</h3>v1 :Still experimenting with Z-image Base training. Showcase images :<ul><li>Euler / normal / 20 steps / CFG 5</li><li>Hires steps : 20 / NMKD_SIAX @0.2-0.5 / x1.5-x2</li><li>768x1152</li><li>Lora strength 0.7 -1</li></ul><h3 id="qwen">Qwen</h3>Qwen2512 v1:First lora trained on the new Qwen2512 (it still working with previous Qwen-Image and Qwen-Edit). For inference, I still prefer the first Qwen-Image model (BF16) (all the showcase images are done with this one).Seems pretty good but few things appears sometimes unprompted in my tests: artefacts on skin (like scars) and japanese onomatopoeia Qwen v1:I've trained a lot of different versions for this one : with big and small dataset, trigger words or not, high and low LR... I've still no idea what's good or not. 🙂 <h3 id="flux">Flux</h3>Had good results starting prompt with : 'black and white Berserk manga style image of'and 0.8 strength most of the time v2 edit : Better dataset for better prompt adherence and flexibility

Manga style (Berserk)

Better flexibility and better prompt adherence

berserk_manga_style_flux_v2.safetensors

Flux v2

Klein 9B v1

ZIB v1

Qwen2512 v1

Qwen v1

Flux v1

Flux :Had good results starting prompt with : 'black-and-white manga berserk style image of'and 0.8 strength most of the time

Files with SHA256: a747490825d9ff92fc14c25af5123010a72ada7ef5a777293762bff06038c745