sinRav | Artist Style LoRA

A style LoRA trained on the artwork of @sinrav on X/Twitter.

All credit for the style goes to sinrav. If you like the results, please follow and support the artist directly. This LoRA is an unofficial fan project and is not affiliated with or endorsed by them.

Trigger word

sinrav

Recommended settings

• Base model: [Illustrious / NoobAI / Pony / SDXL]

• LoRA weight: 0.7–1.0 (start at 0.8. go lower if it fights your character LoRAs)

• Sampler: Euler a 25–30 steps

• CFG: 5–7

Tips

• It stacks well with character LoRAs at around 0.6–0.8 each.

• If the style comes out too strong, drop the weight before you touch the prompt.

• Some source images carried the artist's signature. If it shows up, add signature, artist name to the negative prompt.

Training details

• Images: 114 images from sinrav's public posts, cropped and fitted to buckets

• Captions: [WD14 / booru tags], with the trigger word first

Please don't

• Pass off generations as sinrav's own work or sell them as such

• Upload outputs to sinrav's tags or mentions on social media

• Use this to impersonate the artist