PixelMuseAI/Flux1 Dev Consistent Character - Fast Generation (PuLID, Controlnet, Torch Compile)
This workflow generates a flux image with a consistent character at 1.15s/it on a 4060Ti (on 2nd generation onwards, if the compilation of the model is enabled). You will need 16GB VRAM for this workflow. It uses the fp8 flux models because the gguf modes do not work well with torch.compile. Please setup the environment as per the instructions below if you are getting errors. PuLID for character consistency. Please note that users have reported that PuLID works well on AI generated characters but do not work well on real photographs. FLUX.1-dev-ControlNet-Union-Pro controlnet (need to use Q2_k
| Author / organization | PixelMuseAI |
|---|---|
| Downloads | 3K (as reported by Civitai) |
| Likes | 82 |
| Task | text-to-image |
| Created | 6 November 2024 |
| Last updated at source | — |
Popularity, last 2 days
3,036 now. Counts as reported by the source, snapshotted daily by Silytics.
More from PixelMuseAI
- PixelMuseAI/Flux Inpaint and Outpaint WorkflowModel
- PixelMuseAI/LTX-2.5 Image Audio to Video - LipsyncModel
- PixelMuseAI/Tik Tok Dance Workflow (Mimic Motion + Animate Diff)Model
- PixelMuseAI/Flux Controlnet UpscaleModel
- PixelMuseAI/Tiled Diffusion Tile Controlnet UpscaleModel
- PixelMuseAI/Tik Tok Dance Workflow (Animate LCM, IPAdapter, Controlnet)Model
Silytics shows public metadata from Civitai and links to the original. Information may be incomplete or out of date. Last refreshed 16 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.