About this model
Flux 2 Pro is the production-tier text-to-image and editing model in Black Forest Labs' Flux 2 family, released on November 25, 2025. It sits alongside its larger sibling Flux 2 Max and the editing-focused Flux 2 Max (Edit), with the broader lineup spanning managed APIs and open-weight checkpoints. Black Forest Labs positions Flux 2 Pro as a fast, precise workhorse for photorealistic output, complex scene composition, and reliable prompt adherence.
Architecturally, Flux 2 builds on a latent flow matching approach that combines generation and editing in a single design. The model couples the Mistral-3 24B parameter vision-language model with a rectified flow transformer—the VLM contributes real-world knowledge and contextual understanding, while the transformer handles spatial relationships, material properties, and compositional logic. It supports multi-reference input for maintaining character, product, and style consistency across outputs.
Compared with the earlier Flux 1 generation, Black Forest Labs describes Flux 2 as capturing compositional logic that earlier architectures could not render, with improved text rendering for typography, infographics, and UI layouts that previously suffered from garbled letters.
Black Forest Labs reports running third-party safety evaluations on the Flux 2 checkpoints, including filtering and fine-tuning to mitigate generation of unlawful content before release. As the company frames it, Flux 2 Pro targets prompt adherence and visual fidelity while generating images faster and at lower cost than its own prior tiers.
This About section is AI-generated from public sources (Claude Opus 4.8), with no human editing. It may contain inaccuracies — verify critical details against the sources listed above.
Data sources: Venice API · HuggingFace · Wikipedia — enrichment updated 1d ago