# FLUX 1 Schnell > FLUX 1 Schnell is available on Artifio.ai — a unified AI workspace with a single pay-as-you-go wallet. - Canonical page: https://artifio.ai/models/bfl-flux-1-schnell - Run it: https://artifio.ai/create/bfl-flux-1-schnell - Modality: prompt to image - Model brand: Black Forest Labs - Pricing: from about $0.002 per generation (exact cost shown in the workspace before you confirm; varies with selected options) - Typical generation time: ~8s (catalogue estimate) - Full pricing table: https://artifio.ai/pricing/models ## About FLUX 1 Schnell FLUX 1 Schnell is the fastest member of the FLUX.1 text-to-image family that Black Forest Labs launched on 1 August 2024. The company was founded in Freiburg, Germany by Robin Rombach, Andreas Blattmann and Patrick Esser, researchers who previously worked on image generation at Stability AI and LMU Munich. All three FLUX.1 launch variants share a 12-billion-parameter rectified flow transformer; Schnell is the one distilled for speed and released under the permissive Apache 2.0 license. The name is German for fast. The model is trained with latent adversarial diffusion distillation, which lets it produce a finished image in 1 to 4 sampling steps instead of the dozens a conventional diffusion model needs. On Artifio it runs in the same workspace and wallet as 100+ other models, paid per generation. ## Features - 12-billion-parameter rectified flow transformer built from multimodal and parallel diffusion transformer blocks with rotary positional embeddings - Generates images in 1 to 4 sampling steps thanks to latent adversarial diffusion distillation - Supports a wide range of aspect ratios and resolutions between 0.1 and 2.0 megapixels - Released under the Apache 2.0 license, which permits personal, scientific and commercial use - Prompt adherence that Black Forest Labs describes as competitive with larger non-distilled models ## What people use it for - Rapid concept exploration where many prompt variations are generated and compared in quick succession - High-volume asset production such as thumbnails, placeholders and mood boards - Prototyping a prompt cheaply before re-running the strongest candidates on a slower, higher-quality model - Projects that need an openly licensed model, since Apache 2.0 places no usage restrictions on the weights ## Reported strengths - The 1-to-4-step design makes it one of the quickest routes from prompt to usable image in the FLUX.1 family - It was the most permissively licensed FLUX.1 launch variant, which made it a common default for open tooling - Press testing of the FLUX.1 launch family reported prompt fidelity comparable to DALL-E 3 and photorealism close to Midjourney v6, with notably better hand rendering than Stable Diffusion XL ## Reported limitations - Black Forest Labs made its launch benchmark claims of surpassing Midjourney v6.0 and DALL-E 3 HD for the pro and dev tiers; Schnell is described as competitive with larger non-distilled models rather than ahead of them - The official model card states it cannot reliably render factual information and, as a statistical model, may reproduce societal biases from its training data - Prompt following is strongly influenced by prompting style, so output quality is uneven across phrasings - Black Forest Labs has not disclosed what the model was trained on, and reporters have raised questions about scraped training data ## Sources 1. Announcing Black Forest Labs (Black Forest Labs): https://bfl.ai/announcements/24-08-01-bfl 2. FLUX.1 [schnell] model card (Black Forest Labs (Hugging Face)): https://huggingface.co/black-forest-labs/FLUX.1-schnell 3. FLUX.1 (Wikipedia): https://en.wikipedia.org/wiki/FLUX.1 --- About Artifio.ai: 100+ AI image, video, and audio models in one workspace with a single shared wallet — no per-model subscriptions. This is the markdown rendition of https://artifio.ai/models/bfl-flux-1-schnell for AI assistants and crawlers; the HTML page is canonical.