FLUX AI Generator
Generated on PixelDojo with FLUX (Flux Dev). Produced by PixelDojo's generation pipeline.
FLUX defaults to the Flux Dev variant at 1 credit per image, and our portrait test came back in about 21 seconds on August 26, 2026. The FLUX entry point on PixelDojo is a single API model id that dispatches across Schnell, Dev, Pro, Kontext, and several LoRA remix variants, each priced differently. This page covers what the default Dev variant costs, how fast it ran, what it returned, and where a pricier or cheaper sibling makes more sense.
The 1 credit portrait, Flux Dev, first result
Every example below was produced on PixelDojo. Hover to see the prompt.
Portrait of a middle-aged fisherman mending a net at golden hour, weathered hands in focus, shallow depth of field, natural skin texture, documentary photography style
FLUX (Flux Dev)
Where Flux Dev sits and what runs next to it
Switch models without switching tools. Each one runs in the same PixelDojo studio.
How the FLUX entrypoint actually dispatches
One model id, many variants
The request takes a model field naming the exact FLUX variant identifier, covering Schnell, Dev, Pro, 1.1 Pro, Kontext Pro and Max, and several LoRA remixes. Omit the field and the schema defaults to the Dev variant, which is what we ran.
What our portrait returned
A weathered older man against a harbor sunset, gripping a thick rope with one hand rather than the net our prompt asked him to be mending. His face, beard, and the backlit water read well, but the prompt's specific hand and net detail did not land as literally as it did on other models we have tested with the same wording.
Pricing is per variant, not one flat number
Flux Schnell is free for one image and scales up from there. Flux Schnell LoRA is 0.5 credits per image. Flux Kontext Pro is 1 credit and Kontext Max is 2. Everything else, including the Dev variant we tested, is 1 credit per output by default.
Eleven aspect ratios plus a custom mode
1:1, 16:9, 9:16, 21:9, 9:21, 3:2, 2:3, 4:5, 5:4, 3:4, and 4:3, plus a custom width and height option. Note that Flux Pro and 1.1 Pro reject the two ultra-wide ratios and fall back to 16:9 or 9:16.
LoRA support on specific variants
hf_lora, hf_loras, lora_weights, and a customLoras array are all accepted, but only the LoRA-capable variants such as flux-dev-lora and flux-schnell-lora actually apply them.
One prompt, default variant, 1 credit, about 21 seconds. Same day, Flux 2 Pro ran the identical prompt for 1.5 credits.
Why Choose Pixel Dojo for FLUX
Professional-quality results with cutting-edge AI technology
1 credit per image on the default variant
Flux Dev, the default when model is omitted, is 1 credit per output. Switch the model string for a different price tier.
About 21 seconds, measured
Timed on August 26, 2026 from request to finished file. It returned inside the API's synchronous window.
Strong face and light, looser on small object detail
The backlit harbor scene and the man's face composed well. The rope in his hand did not match the net our prompt described, which is worth knowing before you rely on the model for a very specific small prop.
How It Works
How we ran the 1 credit portrait test:
Leave the model field on its default
Omitting model routes to the Dev variant at 1 credit per image. Naming a different variant identifier changes both the price and the look.
Describe the scene and the detail that matters most
Our prompt named a fisherman mending a net, weathered hands, and shallow depth of field. The face and light landed; the net detail drifted to a rope.
Generate in the app or over the API
From the Flux Studio tool page, or POST to /api/v1/models/flux/run. Ours charged 1 credit and returned in about 21 seconds.
Loved by creators on PixelDojo
Real feedback from people using PixelDojo, pulled from our in-product surveys.
amazing web site
phenomenal site. would highly recommend
Amazing features, easy to use, privacy
the number of options, and especially the quick response to questions on Discord
Love you guys!!
Trained my Lora super fast. Still working out how to creat content wit it, but I love it so far.
Explore more AI tools on PixelDojo
AI Tools
Compare & Switch
- Best AI Image Generators
- Best AI Video Generators
- Midjourney Alternatives
- Civitai Alternatives
- Runway Alternatives
- Leonardo Alternatives
- Pika Alternatives
- Luma Alternatives
- Magnific Alternatives
- Veo Alternatives
- Flux Alternatives
- Freepik Alternatives
- Seedance Alternatives
- Seedream Alternatives
- Pixverse Alternatives
- GPT Image Alternatives
- Synthesia Alternatives
- Playground Alternatives
- NightCafe Alternatives
- Canva AI Alternatives
- ElevenLabs Alternatives
- ComfyUI Alternatives
- Fal Alternatives
- Replicate Alternatives
Common Questions
Everything you need to know about FLUX
How much does FLUX cost?
The default Dev variant is 1 credit per image. Pricing changes by variant: Schnell is free for a single image, Schnell LoRA is 0.5 credits, Kontext Pro is 1 credit, and Kontext Max is 2 credits per image.
How fast is FLUX?
About 21 seconds in our August 26, 2026 run on the default Dev variant, from request to finished file, returned inside the API's synchronous window.
How does Flux Dev compare with Flux 2 Pro on the same prompt?
Flux 2 Pro ran the identical fisherman prompt the same day for 1.5 credits in about 21 seconds and composed the fuller, cleaner face with visible stubble that we were looking for. Flux Dev, at 1 credit, produced a similarly strong face and backlit scene but swapped the net our prompt named for a rope. Pay the extra half credit for Flux 2 Pro when a specific small prop detail has to land.
How does it compare with Z-Image Turbo?
Z-Image Turbo is 0.5 credits and, on this same prompt, returned the most convincing skin and hand texture of any model we have tested, in a fifth of the time of the full Seedream tier. Flux Dev costs twice as much and took roughly the same time, so the choice comes down to which model's specific look you prefer rather than price or speed.
Which FLUX variant should I use for image editing?
The FLUX entrypoint documents Kontext Pro and Kontext Max in its model list, but its own request schema and server logic pass no image field through to those variants. For editing an existing photo with a text instruction, use the flux-edit tool instead, which is built for that.
How do I call FLUX from the API?
POST to /api/v1/models/flux/run with your API key, a prompt, and a model value naming the FLUX variant you want (the exact identifier strings are listed in the API reference). Omit model and the request defaults to Flux Dev. Same key as every other model in the catalog.