Verified: a year ago
SafeTensor
The FLUX.1 [dev] Model is licensed by Black Forest Labs. Inc. under the FLUX.1 [dev] Non-Commercial License. Copyright Black Forest Labs. Inc.
IN NO EVENT SHALL BLACK FOREST LABS, INC. BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH USE OF THIS MODEL.
Who wants a thing like this? Maybe just me. If it's you too - welcome to a weird little club.
Trained on mostly old Midjourney outputs with bizarre noise and terrible rendering, but with a distinctive and beautiful coloring and textures.
I did the same for SDXL, and that model actually adapted something much closer to those old outputs, but Flux, which really, really wants to produce pristine photos, resists a lot harder. Especially the more verbose your prompt gets, the less you will see artifacts. But there is nonetheless still a strangeness to the generated image. Prompting direction I gave for the SDXL model holds the same here.
Keywords: Prompt for 'pinhole photography' or 'toy camera photography' - you will likely get a strong vignette or image border and heavy, noisy texture
Poetic prompts: get abstract with your concepts and imagery - this LoRA will be happy to find something in it
Or you can do nothing special, and see where you end up. Sometimes the effect is absurdly strong, and sometimes quite minimal. Really living up to its naming: chaotic.