Black Forest LabsBlack Forest Labs·🖌️ Inpainting

Flux 2 Max

anonymized
Try on Venice.ai ↗
Quick reference
Flux 2 Max — TLDR
  • 🆕 Editing variant of Black Forest Labs' FLUX.2 Max for inpainting and scene edits.
  • 🎯 Targets high-fidelity transformations and detailed photorealistic image editing.
  • 👁️ Couples a Mistral-3 24B vision-language model with a rectified flow transformer.
  • 📏 Edits images at up to 4 megapixels while preserving detail.
  • 🧠 Preserves facial features, proportions, expressions, and visual identity across edits.
  • 🔧 Supports multi-reference editing and grounded generation with web context.
  • 🌐 Available through Black Forest Labs' API and hosted partner endpoints.
💰 Pricing
$0.190
per edit
📅 On Venice since
Jan 5, 2026
149 days ago
Provider

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…

Read full profile →
3 models on Venice
2 image · 1 inpaint
Since Nov 25, 2025

About this model

Flux 2 Max is the editing-focused configuration of Black Forest Labs' FLUX.2 family, tuned for inpainting, outpainting, and scene edits with photorealistic fidelity. It belongs to the same generation as the image-generation variants Flux 2 Max image and Flux 2 Pro, sharing FLUX.2's unified architecture that combines text-to-image generation and image editing in a single model.

The defining FLUX.2 advance over the prior FLUX.1 generation is architectural. Where FLUX.1 added editing through a separate toolkit—FLUX.1 Fill for inpainting and outpainting, plus Depth, Canny, and Redux controls layered atop the base model—FLUX.2 folds generation and editing into one system. It couples a Mistral-3 24B vision-language model with a rectified flow transformer, so real-world knowledge and contextual understanding inform spatial relationships, material properties, and compositional logic.

Black Forest Labs describes the Max variant as the top-tier editing tier of the FLUX.2 lineup, oriented toward editing consistency and prompt following while preserving facial features, proportions, expressions, and overall visual identity across edits. All FLUX.2 variants support editing from text and multiple references at resolutions up to 4MP.

As the top editing tier, Flux 2 Max also supports grounded generation with real-time web context and multi-reference control, positioning it for marketing, design, and e-commerce workflows that require faithful, repeatable edits.

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