IMAGE
FLUX (Black Forest Labs) on Elevence AI
Black Forest Labs' FLUX models — the strongest open-source image generators, on Elevence AI's image API.
Applies to web, iOS and Android
PROVIDER
FAL
MODALITY
image
PRICING
From $0.025 per image (FLUX.1 schnell) to $0.075 (FLUX.1 pro)
FLUX is the image-generation family from Black Forest Labs (the team behind Stable Diffusion). FLUX.1 pro is the highest-quality model — sharp, photorealistic, with state-of-the-art prompt adherence. FLUX.1 dev and schnell are faster, cheaper variants for iteration and high-throughput pipelines.
On Elevence AI you can call any FLUX variant through the same image-generation API as DALL·E, SDXL, Ideogram and Leonardo. Switch between them by changing the model identifier — same request shape, same credit pool.
What FLUX (Black Forest Labs) is great at
State-of-the-art prompt adherence
FLUX.1 pro follows detailed prompts more reliably than DALL·E 3 or SDXL on most tests — text-in-image, complex composition, specific subject attributes. Particularly strong for "render exactly what I asked for" workflows.
Photorealistic quality at production speed
Pro outputs are sharp, with realistic lighting and texture. Schnell variant generates in ~1 second per image — good for previews and high-throughput pipelines where pro-tier quality is unnecessary.
Open-weight ecosystem
FLUX dev weights are publicly available, which means a healthy fine-tune and LoRA ecosystem exists. If your brand has a custom FLUX LoRA, you can typically run it via the same API.
Best for
•
High-quality marketing visuals and ad creative
•
Photorealistic product mockups
•
Concept art and storyboarding
•
High-throughput preview pipelines (schnell variant)
Limitations
Hands and faces still imperfect
Like all diffusion image models, FLUX occasionally produces malformed hands or off-anatomy faces. Newer models (DALL·E 3, Ideogram v3) are sometimes more reliable here. For human-heavy outputs, generate multiple candidates and pick the best.
Pricing higher than SDXL or Stable Diffusion 3.5
FLUX.1 pro is among the more expensive image-generation models. For lower-stakes generations where SDXL or SD 3.5 quality suffices, those are 3–5× cheaper.
Code example
Drop-in OpenAI-compatible API. Swap the base URL and key, change the model field:
// On the Elevence AI studio, pick FLUX.1 pro from the model picker.
// Or via the REST API:
const response = await fetch('https://api.elevence.ai/api/v1/generate/image', {
method: 'POST',
headers: {
'Content-Type': 'application/json',
'Authorization': `Bearer ${process.env.ELEVENCE_API_KEY}`,
},
body: JSON.stringify({
model: 'flux-1-pro',
prompt: 'A studio portrait of a cat wearing wireframe glasses, 35mm film grain',
aspect_ratio: '3:4',
}),
});
Frequently asked questions
Which FLUX variant should I use?
Default to FLUX.1 dev — best price/quality balance. Upgrade to FLUX.1 pro for final production assets where every detail counts. Use FLUX.1 schnell for fast previews and high-throughput pipelines where 1-second generation matters more than peak quality.
How does FLUX compare to DALL·E 3?
FLUX.1 pro and DALL·E 3 are competitive at the top end. FLUX tends to be slightly better at photorealism and detailed composition; DALL·E often wins on text rendering and human anatomy. For brand-consistent visuals, also try Ideogram.
Can I use my custom FLUX LoRA on Elevence AI?
Custom LoRA support is in development. Today, you can use the base FLUX models. Reach out if you have a specific LoRA you'd like to deploy — we're prioritising based on customer demand.
Does Elevence AI generate images in batches?
Yes. Request multiple images per call (typically 1–4) for variation. The pricing is per-image generated; batch is just a convenience.
What aspect ratios are supported?
Standard ratios: 1:1, 16:9, 9:16, 3:2, 2:3, 3:4, 4:3, 4:5, 5:4. Custom dimensions also work within the model's output-size limits.
Related models
Try FLUX (Black Forest Labs) on Elevence AI
Sign up free and access FLUX (Black Forest Labs) alongside 60+ other models — one bill, pay-as-you-go.
Get started →