Image generation and editing

How to use
FLUX.

Creating and editing high-fidelity images with FLUX.2 variants, including multi-reference composition, typography, precise colors, and production image workflows.

What it isBlack Forest Labs image generation and multi-reference editing model platform Workflows2 UpdatedSeptember 2026

FLUX is a Black Forest Labs image generation and multi-reference editing model platform. Creating and editing high-fidelity images with FLUX.2 variants, including multi-reference composition, typography, precise colors, and production image workflows. This guide covers the whole path in one place: official access, a first session that produces something reviewable, the checks that make output trustworthy, and the permissions worth limiting before you connect real work.

Troiana principle

AI should make the work easier to inspect. If the workflow removes the source, the owner, or the review step, redesign the workflow.

01

Access & setup

Find, install, and sign in to FLUX

Get into the official FLUX experience with the right account and a setup you understand.

  1. 01

    Start at https://bfl.ai/ and confirm the domain before entering account or payment information.

  2. 02

    Availability: FLUX models are available through the BFL playground and API, supported partners, and selected downloadable weights with model-specific licenses.

  3. 03

    A BFL account and project, credits and project-scoped API key for hosted use, or compatible hardware and the exact applicable license for local weights.

  4. 04

    Sign in with the account you intend to keep using, then review plan, data, notification, and permission settings.

  5. 05

    Run one low-risk test task before connecting sensitive files, repositories, or workspace data.

  • Use official download pages.
  • Review permissions during setup.
  • Keep installers and applications updated.
02

First session

Your first useful FLUX session

Learn the interaction loop using a small task with a clear outcome.

  1. 01

    Create a development project and key, keep the key server-side, and add a controlled credit amount.

  2. 02

    Choose a fixed FLUX.2 endpoint based on quality, speed, controls, resolution, and license needs.

  3. 03

    Submit a low-risk prompt with explicit dimensions and record the endpoint, request ID, settings, and expected cost.

  4. 04

    Poll with bounded retries, retrieve the signed result promptly, and inspect quality and rights concerns before retention.

  • State the outcome before the background.
  • Provide the real source material.
  • Review the result before expanding the task.
03

Quality control

Check the quality of FLUX output

Turn a promising generation into an asset you can legally and safely ship.

  1. 01

    Restate the brief, the deliverable format, and where the asset will appear before judging any output.

  2. 02

    Inspect at full resolution: generated media hides its errors at thumbnail size.

  3. 03

    Check the details these systems reliably get wrong — hands, text, logos, reflections, symmetry, and edge artefacts.

  4. 04

    Confirm the licence and commercial-use terms for your plan, and check the output does not reproduce a recognisable style, mark, or likeness you have no right to use.

  5. 05

    Review for accessibility and delivery: alt text, contrast, colour profile, file size, and the exact export format.

  • Inspect at full resolution before approving anything.
  • Confirm commercial-use rights before an asset reaches a client.
  • Keep the prompt and settings with the asset so it can be reproduced.
04

Privacy & permissions

Use FLUX safely

Use the platform without surrendering rights, likenesses, or client material you do not control.

  1. 01

    Confirm you hold the rights to every reference image, voice, or clip before uploading it.

  2. 02

    Check whether your plan trains on submitted content, and opt out before uploading client work.

  3. 03

    Never upload a real person's likeness or voice without their explicit, documented consent.

  4. 04

    Review the output for marks, styles, or likenesses you have no licence to reproduce.

  5. 05

    Label synthetic media where your client, platform, or jurisdiction requires disclosure.

  • FLUX variants differ in endpoint stability, controls, pricing, hardware needs, and licenses; preview endpoints can change. Protect project keys, use fixed endpoints for reproducibility, verify every weight license, confirm rights and consent for references, minimize uploaded assets, prevent duplicate asynchronous requests, retrieve expiring results promptly, inspect identity and trademark changes, and retain request provenance and human approval.
  • Follow your organisation's approved-use policy.
  • Never treat fluent output as authorization to act.
05

Core workflows

Step-by-step ways to use FLUX for the work it does best.

Each workflow below has its own outcome, steps and review checkpoints.

Workflow 01

Choose and benchmark a FLUX.2 image model

Compare fixed FLUX.2 variants for generation quality, control, speed, licensing, and megapixel-based cost.

Outcome: Select a reproducible production endpoint without depending on a changing preview model or a mismatched license.

  1. 01

    Define the visual task, target resolution, references, typography and color needs, latency, volume, rights constraints, and evaluation rubric.

  2. 02

    Review FLUX.2 variant capabilities, API pricing, project isolation, open-weight licenses, and fixed versus preview endpoint behavior.

  3. 03

    Create a project-scoped key in a secret store and run the same versioned prompts on shortlisted fixed endpoints with explicit output settings.

  4. 04

    Measure prompt adherence, identity consistency, text, color, artifacts, diversity, latency, result failures, and cost per accepted image.

  5. 05

    Pin the chosen endpoint and settings, save request IDs and seeds where available, retrieve signed results promptly, and regression-test upgrades.

What good use looks like

  • Prefer fixed endpoints for reproducibility.
  • Check each model's license.
  • Measure cost per accepted image.

Workflow 02

Create a controlled multi-reference edit with FLUX.2

Combine authorized references while specifying each image's role and checking identity, branding, composition, and unintended copying.

Outcome: Generate a traceable composite that uses only approved inputs and preserves the intended elements without hidden substitutions.

  1. 01

    Confirm rights and consent for every reference, label each image's intended role, and define protected identity, product, logo, text, and layout details.

  2. 02

    Choose a FLUX.2 variant and output size within reference and megapixel limits, then upload only the minimum necessary sources.

  3. 03

    Write an explicit prompt mapping each reference to subject, pose, product, style, or structure and state what must remain unchanged.

  4. 04

    Poll the asynchronous request with bounded retries, retrieve the short-lived result immediately, and handle failed states without duplicate charges.

  5. 05

    Compare every reference and output at full resolution, reject unintended identity or trademark changes, and archive inputs, prompt, model, and approval.

What good use looks like

  • Assign every reference a clear role.
  • Minimize uploaded source images.
  • Retrieve signed results promptly.
06

Official references

Check the current product documentation.

Features, plan limits, availability, and data controls change. These official pages are the starting points used for this guide.

Explore the Troiana AI Hub →