is now Verda Learn more
FLUX.2 by Black Forest Labs
Enterprise discounts available - Talk to sales

FLUX.2 by Black Forest Labs

SOTA image generation and editing

at $0.015 / output

Faster inference speed Lower token cost No loss in model quality

FLUX.2 models & pricing

Secure access and easy integration with API
  • Black Forest Labs Verda WaveSpeedAI

    FLUX.2 [dev]

    The new base model from Black Forest Labs, trained to set a new standard for image generation and image editing.

    An ultra-fast and cost-efficient inference endpoint by Verda - ensuring secure access and seamless integration with API.

    Cost per image:

    $0.015

  • Black Forest Labs Verda

    FLUX.2 [pro]

    A variant of FLUX.2 that emphasizes output quality over inference speed, delivering the highest possible quality in image generation and editing.

    Available via the Verda Cloud Platform, running on the Black Forest Labs infrastructure.

    Cost per image:

    $0.09

  • Black Forest Labs Verda

    FLUX.2 [flex]

    A variant of FLUX.2 that offers the highest output quality and character consistency at the expense of higher latency.

    Available via the Verda Cloud Platform, running on the Black Forest Labs infrastructure.

    Cost per image:

    $0.40

FLUX.2 model overview

Beyond pixel generation toward visual intelligence
A pioneer for iterative image editing

FLUX.2 is a new base model by Black Forest Labs, raising the bar for image generation and editing with new capabilities (like multi-ref images), higher resolution (up to 4MP), and the highest quality to date.

Text-to-image prompting with FLUX.2 gives a high degree of control over different aspects of output, such as subject, action, style, and context. FLUX.2 is trained to understand structured JSON prompts and allows the use of HEX codes to describe object color palettes.

FLUX.2 similarly advances image-to-image editing. The model's core capabilities cover a wide range of editing tasks, from composite edits with reference inputs and natural language to precise edits such as color change.

Get API and prompting guides from Black Forest Labs.
Optimized for large-scale, production-grade inference

Managed Endpoints by Verda

Fast Run models faster and cheaper without loss in quality
Scalable Handle dynamic usage patterns with zero engineering overhead
Secure Get secure API access with bearer token authentication
Cost-effective Pay per output and manage billing from the dashboard

Looking for something different?

Other inference services

  • Serverless Containers

    Create fast and scalable endpoints for containerized models of your choice. Leverage queue-based scaling, pay per use, and scale to zero.

  • Co-development

    Our engineering team can be your strategic partner for building and maintaining custom inference solutions, tailored to your use cases.

You can also reach us via the contact form