--- title: "FLUX Schnell: Specs, Speed & Prompt Tips | imaginary.app" description: "FLUX.1 [schnell] is Black Forest Labs’ few-step, Apache 2.0 image model. What it is good at, where it falls short, and how to prompt it for quick drafts." url: https://www.imaginary.app/models/flux-schnell --- Updated November 2026 # FLUX Schnell (FLUX.1 \[schnell]) FLUX.1 \[schnell] is the speed-first member of Black Forest Labs’ original FLUX.1 family: a distilled model that produces an image in 1 to 4 steps. It suits fast concept drafts and bulk ideation more than final, detail-critical artwork. Key takeaways - A 12-billion-parameter rectified flow transformer distilled to generate in 1 to 4 steps, per Black Forest Labs’ model card. - Released August 1, 2024 as the fastest of the three FLUX.1 variants; it is the oldest model in this image lineup. - Openly licensed under Apache 2.0, which is unusual among the image models we offer. - Use it to explore many compositions quickly, then move a chosen idea to a slower, higher-fidelity model such as FLUX.2 Pro. Black Forest Labs Generate with FLUX Schnell Write a prompt, pick FLUX Schnell, choose your settings and see the credit cost before you start. Failed generations return their credits. [Open the generator](https://www.imaginary.app/generate)[View Pricing](https://www.imaginary.app/pricing) Sign in with Google to create an account; buy credits only when you need them. ## On imaginary.app These are the settings our generator offers for FLUX Schnell right now, read from our live model catalog. - Type Image, from a text prompt - Resolutions 1K - Aspect ratios 16:9, 9:16, 1:1 - Starting price 0.07 credits (about $0.01–$0.02) at the lowest settings (1K, 16:9) The generator shows the exact credits for your settings before you start. USD ranges depend on which [pack or plan](https://www.imaginary.app/pricing) you bought credits with. If a generation fails, or you cancel it before it starts, its credits go back to your balance. Very rarely, when we can't confirm with the provider whether a generation finished, the credits stay reserved while we check, rather than being charged. Models other than Veo 3.1 need an account that has bought a credit pack or plan, and they have daily usage limits, per account and across the site: if one is reached, the generator tells you and you can try again the next day (limits reset at midnight UTC). ## Built for speed through distillation Most diffusion-style image models refine an image over dozens of steps. Black Forest Labs trained Schnell with latent adversarial diffusion distillation so that it reaches a usable image in 1 to 4 steps (“schnell” is German for “fast”). The trade-off is that the model has less room to correct itself, so fine detail, small objects and complex multi-subject scenes are less dependable than in larger, slower models. Think of it as a sketchpad: quick to fill, cheap to throw away. ## Open weights under Apache 2.0 Schnell’s weights are published under Apache 2.0, a permissive licence, and the model card states it can be used for personal, scientific and commercial purposes. That matters if you want to prototype here and later run the same open model yourself, for example with Hugging Face Diffusers. ## An older generation, and it shows in text and detail Schnell launched in 2024. Black Forest Labs presents reliable fine text, infographics and UI mockups as an improvement of its FLUX.2 generation (November 2025), and it has since announced FLUX 3 (July 2026). In practice that means lettering in Schnell is hit-and-miss: one or two large words can work, but paragraphs, small labels and logos usually come out garbled. If words in the image matter, pick a typography-focused model instead. ## FLUX Schnell (FLUX.1 \[schnell]) specs from Black Forest Labs What Black Forest Labs and its API partners document for the model itself. A maker spec is not always a setting our generator offers: what you can choose on imaginary.app is listed in the block above. | Spec | As documented | Source | | ------------------- | -------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------- | | Release | August 1, 2024, as part of the FLUX.1 suite (\[pro], \[dev], \[schnell]) | [Black Forest Labs — Announcing Black Forest Labs (FLUX.1 launch, August 1, 2024)](https://bfl.ai/announcements/24-08-01-bfl) | | Architecture | 12-billion-parameter rectified flow transformer | [Black Forest Labs — FLUX.1 \[schnell\] model card (Hugging Face)](https://huggingface.co/black-forest-labs/FLUX.1-schnell) | | Steps | 1 to 4 steps; distilled with latent adversarial diffusion distillation | [Black Forest Labs — FLUX.1 \[schnell\] model card (Hugging Face)](https://huggingface.co/black-forest-labs/FLUX.1-schnell) | | Positioning | The fastest FLUX.1 model, “tailored for local development and personal use” | [Black Forest Labs — Announcing Black Forest Labs (FLUX.1 launch, August 1, 2024)](https://bfl.ai/announcements/24-08-01-bfl) | | Resolution range | FLUX.1 variants support aspect ratios and resolutions between 0.1 and 2.0 megapixels | [Black Forest Labs — Announcing Black Forest Labs (FLUX.1 launch, August 1, 2024)](https://bfl.ai/announcements/24-08-01-bfl) | | Licence | Apache 2.0; usable for personal, scientific and commercial purposes per the model card | [Black Forest Labs — FLUX.1 \[schnell\] model card (Hugging Face)](https://huggingface.co/black-forest-labs/FLUX.1-schnell) | | Safety filter (fal) | Images flagged by fal’s safety checker are returned as a black image | [fal — FLUX.1 \[schnell\] API schema](https://fal.ai/models/fal-ai/flux/schnell/api) | | Successor | FLUX.2, released November 25, 2025 | [Black Forest Labs — FLUX.2 announcement](https://bfl.ai/blog/flux-2) | | Newer family | FLUX 3, announced July 23, 2026 | [Black Forest Labs — FLUX 3 announcement](https://bfl.ai/blog/flux-3) | ## Who FLUX Schnell (FLUX.1 \[schnell]) is for ### Good fit - Brainstorming: trying ten framings of an idea before committing to one. - Mood boards, thumbnails and rough storyboards where speed beats polish. - Simple subjects: a single object, a landscape, a stylised character on a plain background. - Testing prompt wording cheaply before running it on a premium model. ### Look elsewhere if - Posters, logos, menus or anything that needs accurate words in the image. - Final product shots or photoreal portraits where skin, hands and small details are scrutinised. - Dense scenes with many named objects that must all appear in the right place. ## Prompt tips for FLUX Schnell (FLUX.1 \[schnell]) Keep prompts short and concrete. The model card warns that prompt following is heavily influenced by prompting style; with only a few steps, a clear subject, setting and style in plain words works better than a long list of adjectives. Example prompt (written by us) ``` A red vintage bicycle leaning against a whitewashed wall, Mediterranean afternoon sun, soft shadows, 35mm photo ``` Lead with the subject and put style last. With only a few steps there is little room to recover a subject buried behind lighting jargon, so name the main thing first. Example prompt (written by us) ``` Isometric illustration of a tiny coffee shop on a floating island, pastel colours, clean flat shading ``` Describe what you want rather than what you don’t. There is no negative-prompt field here, and words like “no people” can make the model think about people. Use it as a draft stage. When a composition works, reuse the same prompt on FLUX.2 Pro or Seedream 4.5 for a cleaner, more detailed final. Example prompt (written by us) ``` Moody concept art of a lighthouse on a cliff during a storm, wide shot, dramatic clouds, painterly style ``` ## Limitations - Text inside images is unreliable beyond one or two large words. - Fine detail (hands, jewellery, small background objects) is weaker than in newer, slower models. - Black Forest Labs’ model card notes that the model may fail to match the prompt and is not intended to provide factual information. - On imaginary.app this model is offered for text-to-image only; image editing and other FLUX tools are not available here. - Outputs flagged by the safety checker come back as a black image instead of an error. - FLUX Schnell is made by Black Forest Labs; imaginary.app is not affiliated with Black Forest Labs. ## Related models - [FLUX.2 Pro](https://www.imaginary.app/models/flux-2-pro) - [Seedream 4.5](https://www.imaginary.app/models/seedream-4-5) - [Nano Banana 2](https://www.imaginary.app/models/nano-banana-2) - [Every model on imaginary.app](https://www.imaginary.app/models) - [AI image generator overview](https://www.imaginary.app/ai-image-generator) ## Frequently asked questions ### Is FLUX Schnell free? The model weights are free to download under Apache 2.0, so you can run FLUX.1 \[schnell] on your own hardware. Running it on imaginary.app uses credits; the current starting cost is shown in the “On imaginary.app” block on this page. ### FLUX Schnell vs FLUX.2 Pro: which should I use? Use Schnell when you want many quick drafts and can accept rougher detail. Use FLUX.2 Pro for finished images: it is a newer generation (November 2025) that Black Forest Labs positions for production quality, stronger prompt adherence and legible fine text. ### What resolution does FLUX Schnell support? Black Forest Labs says FLUX.1 models support resolutions between 0.1 and 2.0 megapixels across a range of aspect ratios. The exact sizes and aspect ratios available on imaginary.app are listed in the block on this page. ### Can FLUX Schnell write text in images? Short, large words sometimes come out right, but it is not dependable for signs, posters or logos. For typography, Ideogram 4.5, Recraft V4.1 or Nano Banana Pro are better suited. ### Can I use FLUX Schnell images commercially? The model card states FLUX.1 \[schnell] is released under Apache 2.0 and can be used for personal, scientific and commercial purposes. For images generated on imaginary.app, also check our Terms of Service. ## Sources Prices, model specs and dates on this page were checked on November 10, 2026. Third-party prices change; check the source before you rely on a number. - [Black Forest Labs — FLUX.1 \[schnell\] model card (Hugging Face)](https://huggingface.co/black-forest-labs/FLUX.1-schnell) - [Black Forest Labs — Announcing Black Forest Labs (FLUX.1 launch, August 1, 2024)](https://bfl.ai/announcements/24-08-01-bfl) - [fal — FLUX.1 \[schnell\] model page](https://fal.ai/models/fal-ai/flux/schnell) - [fal — FLUX.1 \[schnell\] API schema](https://fal.ai/models/fal-ai/flux/schnell/api) - [Black Forest Labs — FLUX.2 announcement](https://bfl.ai/blog/flux-2) - [Black Forest Labs — FLUX 3 announcement](https://bfl.ai/blog/flux-3) - [imaginary.app pricing](https://www.imaginary.app/pricing) How we checked: model specs come from Black Forest Labs's and its API partners' pages listed above; the imaginary.app settings and starting price come from our live model catalog. Example prompts are ours, not Black Forest Labs's. Commercial use: licence and usage terms for FLUX Schnell (FLUX.1 \[schnell]) are set by Black Forest Labs. Check them, and our [Terms of Service](https://www.imaginary.app/terms-of-service), before you publish what you make. Black Forest Labs ## Try FLUX Schnell on your own prompt One account and one credit balance for FLUX Schnell, every other image model we offer, and Veo 3.1 video. The price shows before you generate. [Open the generator](https://www.imaginary.app/generate)[View Pricing](https://www.imaginary.app/pricing) Sign in with Google to create an account; buy credits only when you need them.