With the FLUX skill GENI generates images from a description, with four models that let you trade speed against quality.
Available from Growth. No connection needed.
What can GENI do with FLUX?
- Generate an image from a text description, with a choice of four models
The model choice decides almost everything: 20 credits for a quick sketch against 1,000 for a premium edit is a factor of 50.
| Model | For | Credits |
|---|---|---|
| Schnell | Fast iterations, concepts, trying things out | 20 |
| Dev | High quality, sharp detail, portraits | 175 |
| Kontext Pro | Adjusting an existing image, context-aware | 500 |
| Kontext Max | Premium context-aware editing | 1,000 |
Example: what you ask, what you get
Make a hero image for my website: an installer at work in a Dutch living room,
warm daylight
Done with Dev (175 credits).
The image is in your Workspace under Files. If you want variations, I can make three with Schnell for 60 credits total; then you pick the direction and have that one redone in Dev.
That is usually the cheapest way to work: explore cheaply first, then render once at quality.
Requirements
- Plan: Growth and up
- Connection: none
Activating
- Go to Dashboard -> Skills and activate FLUX Image Generation
- Describe what you want to see and optionally name the model
What it costs
See the table above: from 20 to 1,000 credits per image. Without a model choice GENI picks a sensible default and states the cost before starting. See The credit system.
Limits
- Do not expect literal text in the image. For text in an image Nano Banana Pro is a better fit.
- No recognisable people or brand logos. Descriptions asking for those are refused.
- You will not always get what was in your head. Expect a few attempts; that is what Schnell is for.
- Every attempt costs credits, including ones you do not like.
- No editing an existing image except with the Kontext models.
- You are responsible for how you use the result.
Troubleshooting
The result does not match your description. Be more concrete about composition, light and style. "An installer" leaves a lot open; "an installer seen from the side, warm daylight from the left" much less.
It costs more than expected. You were probably on Dev or Kontext. Name Schnell explicitly when you want to explore quickly.
Hands or faces come out wrong. A known weakness of image models. Dev does better than Schnell; otherwise frame the shot differently.
Your request is refused. The description touches on recognisable people, brands or other material that may not be generated.
Frequently asked questions
What is the difference with AI image generation? That runs on Nano Banana and is stronger at text-in-image and business assets. FLUX gives you more control over style plus a cheap fast option. See AI image generation.
Where do the images end up? In your Workspace, Files tab. See Files.
May I use the images commercially? Yes, with the caveat that you remain responsible for what you publish.
Can I have an existing image adjusted? Yes, with Kontext Pro or Max. For scaling and cropping, AI image editing is far cheaper.
Back to Skills marketplace See also: AI image generation · AI image editing · Files On the site: all skills
dGENIX Docs, FLUX Image Generation, updated August 2026