- 🎯 Black Forest Labs' top-tier FLUX.2 model for image generation.
- 👁️ Built for photorealism, strong prompt fidelity, and complex scene composition.
- 🔧 BFL positions it for the family's best editing consistency.
- 🌐 Supports multi-reference editing for consistent characters and products.
- 🆕 Sits above the Pro tier as FLUX.2's quality flagship.
- 🏢 Made by Freiburg-based Black Forest Labs, founded by ex-Stability AI staff.
- 🔒 Safety fine-tuned and filtered against unlawful content before release.
- 📚 Released November 25, 2025.
Black Forest Labs is a generative AI company based in Freiburg im Breisgau, Germany, founded by former members of Stability AI. The lab is best known for developing the Flux family of text-to-image models, which generate images from natural language prompts and have quickly…
Explore 2 more models by Flux (text-to-image model) →Flux 2 Max is the top-tier quality model in Black Forest Labs' FLUX.2 family, aimed at photorealistic generation, strong prompt fidelity, and complex scene composition. According to BFL, it delivers the family's strongest prompt following, faithful style representation, and best editing consistency, preserving colors, lighting, faces, text, and objects across complex edits and changing environments.
Within the lineup, Flux 2 Max sits above Flux 2 Pro, which Black Forest Labs positions as a high-quality, production-grade option, while Max targets the most demanding generation and editing work. FLUX.2 also supports multi-reference editing for consistent characters and products across outputs. A dedicated editing-focused variant, Flux 2 Max edit, followed in January 2026.
FLUX.2 is presented by Black Forest Labs as the next generation of its image models, succeeding the earlier FLUX.1 line with improved prompt adherence and higher fidelity. The Max tier represents the upper end of that generation's quality spectrum.
Black Forest Labs is based in Freiburg im Breisgau, Germany, and was founded by former Stability AI employees. Prior to release, the company applied pre-training data filtering, targeted safety fine-tuning, and third-party evaluation to mitigate generation of unlawful content such as CSAM and NCII.
This About section is AI-generated from public sources via VeniceStats + Venice inference, with no human editing. It may contain inaccuracies.
| Seller | Reputation↓ | Routing | $ / img | Categories | API |
|---|---|---|---|---|---|
| Venice.ai Proxy 0x1f22…18c9 | 100.00 | #2 | $0.045 | image,creative | openai-images |
| ▲ Apex Ant 0x73b4…e736 | 91.01 | #1 | $0.04 | image,media | openai-images |
| D5V1N2 0xd5e7…7be0 | 33.17 | gated | $0.1045 | image,creative,router,fallback | openai-images |
"Best price" and the seller table are live AntSeed catalog data (advertised $/1M tokens — or $ per generated image for unit-billed image models — not settled amounts). Reputation = on-chain trust (0-100). "Routing" = the SDK's default buyer routing (what the VPR desktop app ships with): a trust ≥ 60 gate on the effective reputation, then cheapest-first among routable sellers; live failover state (per-peer cooldowns) is buyer-side runtime and not included. Model knowledge (TLDR, provider, About) via the VeniceStats enrichment layer. Advertised catalog, not the model used in any specific purchase. "Usage on AntSeed" counts only settlements whose buyers share the per-model split on-chain (metadata v2/v3, opt-in), so every usage figure is a lower bound.