# FLUX1.1 [pro]

Faster than the original FLUX.1 Pro. More detail at standard sizes. Same prompt format. FLUX1.1 [pro] is the daily-driver model for digital image generation within the FLUX.1 family, tuned for throughput and per-image cost rather than maximum pixel density.

- **Model ID:** `bfl/flux-pro-1.1`
- **Type:** image
- **Providers:** bfl
- **Context window:** Not applicable
- **Maximum output tokens:** 0
- **Pricing:** $0.04/image
- **Canonical page:** https://vercel.com/ai-gateway/models/flux-pro-1.1

## Supported parameters

- **Parameters:** `seed`
- **Operations:** `text-to-image`
- **Input modalities:** `text`
- **Output modalities:** `image`
- **Aspect ratios:** Not reported
- **Resolutions:** Not reported
- **Durations:** Not reported
- **Frame rates:** Not reported
- **Maximum outputs per request:** Not reported
- **Maximum input images:** Not reported
- **Input image formats:** `jpeg`, `png`
- **Maximum input image size:** Not reported

## Example

```ts
import { experimental_generateImage as generateImage } from 'ai';

const result = await generateImage({
  model: 'bfl/flux-pro-1.1',
  prompt: 'A red balloon on a wooden table.'
});
```

## About

FLUX1.1 [pro] shipped on October 2, 2024 as a targeted refinement of FLUX.1 Pro. The underlying architecture and prompt contract didn't change. The improvements are render speed and output fidelity at standard resolutions.

Render velocity matters for production pipelines. Every queued image completes sooner than on the original Pro, and that margin compounds across thousands of generations. Studios cycling through advertising variants, catalog photography, or social-media assets translate per-image savings directly into shorter pipeline runs.

Prompt fidelity improved alongside speed. Descriptions with multiple subjects, explicit spatial arrangements, and targeted lighting conditions resolve with greater consistency. The FLUX.1 visual signature carries over intact, so calibrated prompts and LoRAs trained on FLUX.1 outputs transfer without rework.

This model generates at standard pixel dimensions. If your workflow requires higher resolution output, raw file formats, or the expanded density needed for print and physical media, use the Ultra variant in the same generation.

## What to consider

Teams migrating from the original FLUX.1 Pro should validate prompt behavior on their specific use cases. Iterative improvements can subtly shift aesthetic outputs.

## When to use

### Best For

**High-volume digital asset production:** Web banners, social cards, product shots, and advertising permutations where standard pixel dimensions suffice and per-image cost governs architecture decisions.

**Smooth FLUX.1 Pro migration:** Swap the model identifier, keep everything else. Same prompt syntax, improved results, quicker turnaround.

**Multi-subject scene composition:** Tighter prompt fidelity means scenes with several objects, spatial cues, and lighting directives land more reliably.

### Consider Alternatives When

**Physical media is the deliverable.** The higher-tier FLUX 1.1 variant provides expanded pixel dimensions and raw file output for print and large-format work.

**Multi-reference conditioning is required.** The FLUX.2 generation supports simultaneous reference images in ways FLUX.1 does not.

**Sub-second interactive previews are needed.** FLUX.2 Klein's distilled architecture targets real-time rendering.

## Best for

- **High-volume digital asset production:** Web banners, social cards, product shots, and advertising permutations where standard pixel dimensions suffice and per-image cost governs architecture decisions
- **Smooth FLUX.1 Pro migration:** Swap the model identifier, keep everything else. Same prompt syntax, improved results, quicker turnaround
- **Multi-subject scene composition:** Tighter prompt fidelity means scenes with several objects, spatial cues, and lighting directives land more reliably

## Consider alternatives

- **Physical media is the deliverable:** The higher-tier FLUX 1.1 variant provides expanded pixel dimensions and raw file output for print and large-format work
- **Multi-reference conditioning is required:** The FLUX.2 generation supports simultaneous reference images in ways FLUX.1 does not
- **Sub-second interactive previews are needed:** FLUX.2 Klein's distilled architecture targets real-time rendering

## Frequently asked questions

### What does FLUX1.1 [pro] cost per image?

Check the pricing panel on this page for today's numbers. AI Gateway tracks rates across every provider that serves FLUX1.1 [pro].

### What pixel dimensions does FLUX1.1 [pro] produce?

Standard FLUX.1 resolutions for this slug. See N/A for dimensions and quality tiers. Workflows that need higher resolution or raw file output should use the Ultra variant in the same generation.

### Do existing FLUX.1 Pro prompts carry over?

Yes. Identical prompt syntax, negative-prompt support, and generation parameters carry over. Spot-check critical prompts since quality refinements may shift outputs subtly.

### How much quicker is it than the original FLUX.1 Pro?

Per-image render time dropped measurably. Exact gains depend on provider infrastructure and requested dimensions. Consult available providers for provider-published benchmark figures.

### Should a new project start here or on FLUX.2 Pro?

If the project requires multi-reference conditioning, VLM-based scene understanding, or 4MP resolution, begin with FLUX.2 Pro. If the priority is efficient, high-volume standard-resolution generation with an established prompt ecosystem, FLUX1.1 [pro] is more economical.

### How do I invoke FLUX1.1 [pro] through AI Gateway?

Pass the identifier `bfl/flux-pro-1.1` to `generateImage` from the AI SDK. The request format is shared across FLUX models on the gateway.

## Links

- [Model page](https://vercel.com/ai-gateway/models/flux-pro-1.1)
- [AI Gateway documentation](https://vercel.com/docs/ai-gateway)
- [Provider model documentation](https://bfl.ai/models/flux-pro)
- [Provider pricing](https://bfl.ai/pricing)


If you encounter incorrect information, can’t find what you need, need a capability Vercel doesn’t currently offer, or have any other feedback to report, please do it by following the instructions at https://vercel.com/feedback.md.