Declared capabilities
- Cfgscale
- 1
- Local Checkpoint
- flux1-dev-fp8.safetensors
- Observed Resolution
- 1344x768
- Sampler
- euler
- Scheduler
- simple
- Steps
- 36
- Text to Image
- true
Model record · Black Forest Labs
Generated locally through the SwarmUI JSON API at the operator-authorized 3090 Ti endpoint. No hosted inference provider or per-image API charge was used.
Vision adherence score 8.0/10. Strong museum-object rendering with navy enamel, brass mechanics, spectacles, laboratory props, and a large key; both eyes glow amber instead of exactly one, and the art-deco fan is only partially/asymmetrically visible.
Vision adherence score 4.0/10. Ten objects: 4 red cubes, 4 glass spheres, 1 brass pyramid, and 1 paper crane. The object classes are recognizable but the cube/sphere counts are both high and the pyramid sits between rather than directly behind the spheres.
Vision adherence score 8.0/10. All required words are spelled correctly with no extra copy, but they are arranged as three visual lines ('MEASURE THE' / 'MAGIC' / 'NOT THE MARKETING') instead of the required exact two-line break.