Black Forest Labs releases Flux 1.1 Pro and API

Black Forest Labs (BFL), a startup founded by the creators of the popular Stable Diffusion AI image generation model that underpins many AI image generation applications and services (such as Midjourney), has announced the release of a new, faster text-to-image model tool called Flux 1.1 Pro, and with it a paid application programming interface (API) on which developers can create third-party applications based on this model (or incorporate them into their existing applications).

This means that a creative tools company can add Flux as an option to its offering if it (and, by extension, its end users) are willing to cover the API costs.

Individual users can access the new Flux 1.1 Pro model not through the Black Forest Labs website, but through partners Together.ai, Replicate, fal.ai and Freepik. Some of these services refer to the model by a different name, such as “Flux Fast”.

No details were immediately provided about the Flux 1.1 Pro training dataset, which is a contentious issue for generative AI companies with the original Stability AI and competitor Midjourney being sued by artists who accuse the companies and others of violating their copyrights by mass scraping and training without consent or compensation for human-made images posted online. One of the key class action lawsuits against Stability AI and Midjourney remains in court.

The news comes on the heels of the success of Flux’s first open-source AI model, which powers Elon Musk’s Grok 2 chatbot with xAI and is available to subscribers of his X social network.

Unlike its predecessor Flux.1, which was open source and free for anyone to download, tune, customize and otherwise use for any commercial or personal use of their choosing, the new Flux 1.1 Pro appears to be Flux 1.0 Pro, paid proprietary offering only. However, it is still available for commercial and enterprise use.

BFL sees the launch of API and Flux 1.1 Pro as major steps in its company’s growth, offering developers and enterprises alike access to powerful and customizable image generation tools.

Flux 1.1 Pro codenamed “Blueberry” takes a new top spot on the leaderboard in the AI ​​image arena

Flux 1.1 Pro is an improvement on the earlier Flux 1.0 Pro model, providing six times faster generation speed, while improving image quality, fast grip and diversity.

It enables a workflow that prioritizes speed without sacrificing quality, generating output three times faster than its predecessor.

Additionally, BFL announced an update to the original Flux 1.0 Pro, doubling its generation speed to improve overall performance.

Flux 1.1 Pro’s performance was tested during its secret debut on Artificial Analysis, an independent third-party benchmarking platform for comparing AI model performance, where the model was tested days before today’s announcement under the codename “blueberry.” (Some incorrectly speculated on X that it was OpenAI testing Sora after o1 LLM testing as “strawberry”).

As of October 1, 2024, Flux 1.1 Pro has the highest ELO score on the platform of 1153, outperforming other generative models in terms of visual fidelity and instantaneous accuracy, including Midjourney 6.1 (ELO score 1100) and Ideogram v2 (score 1108).

The independent ELO benchmark was established in the summer of 2024 by Artificial Analysis co-founder and CEO Micah Hill-Smith and co-founder and product leader George Cameron. Human evaluations of image pairs are used to obtain the results.

For users requiring high resolution, Flux 1.1 Pro will soon support ultra-high resolution images (up to 2k), maintaining precision and speed thanks to upcoming API updates.

BFL API offers developers AI image generation starting at 4 cents per image

Complementing Flux 1.1 Pro is the beta BFL API, which brings BFL’s generative capabilities directly to companies and developers looking to integrate cutting-edge image generation into their own applications.

The API offers advanced customization, allowing users to tailor model selection, resolution, and content moderation to their specific needs. It also provides scalability, making it suitable for projects ranging from small to enterprise level.

The BFL API is available at competitive prices, making it attractive to users looking for high-quality results without excessive costs.

For example, the image generation price of Flux 1.1 Pro is $0.04 per image, while the older Flux 1.0 Pro is available at $0.05 per image.

Developers can start API integration today, and BFL promises continuous improvements as the beta progresses.

The company anticipates that the API will open the door to countless creative applications, especially in industries such as design, advertising and entertainment, where the demand for high-quality AI-generated media continues to grow.

Building on initial great success

Black Forest Labs is no stranger to the spotlight. Just two months earlier, the company had secured $31 million in seed capital led by Andreessen Horowitz (a16z) and backed by high-profile investors such as Brendan Iribe, Michael Ovitz and Garry Tan.

As reported by VentureBeat, the launch of BFL and its predecessor Flux 1.0 was widely seen as a major milestone in the AI ​​community.

BFL co-founders Robin Rombach, Patrick Esser and Andreas Blattmann brought their expertise from Stability AI, the team behind Stable Diffusion, to this new venture, with a vision for more accessible, open-source, generative AI tools.

Flux 1.0, which was available in three variants (Flux 1.0 Pro, Flux 1.0 Dev and Flux 1.0 Schnell), gained early recognition for its 12 billion parameter architecture and ability to match or even exceed the output quality of competing models such as MidJourney and DALL- E

The open nature of these models, especially Flux 1.0 Dev and Flux 1.0 Schnell, have positioned BFL as a key player in the open source vs. proprietary AI debate.

Industry context and competition

Black Forest Labs’ decision to release Flux 1.1 Pro comes at a time of increased competition in the generative AI media space, with many creators looking to leverage text-to-image AI models with image-to-video models such as those from Pika, Runway and Luma.

Midjourney and Ideogram directly compete with Flux in the paid, proprietary text-to-image AI model space, while Stability AI continues to offer both open source and proprietary models under the leadership of former Weta (film special effects) CEO Prema Akkaraju and Hollywood director James Cameron (Titanic, Avatar, Terminator), who recently joined the company’s management board.

Integration with the social media platform signals that generative AI is becoming more accessible to mainstream users, raising the stakes for other players in the field.

What’s next for BFL?

Looking to the future, Black Forest Labs is already working to expand its generative AI capabilities beyond images.

The company has focused on text-to-video systems, which could further strengthen its leading position in the AI-based media space.

If successful, BFL’s expansion into video could further disrupt industries such as advertising, content creation and virtual reality. It also comes as Midjourney is also reportedly working on generative video models and AI hardware.

For now, Flux 1.1 Pro and the BFL API represent a significant advancement in generative technology, offering users faster and more efficient tools without sacrificing quality.

Whether it’s its own API or partner platforms like Together.ai, Replicate, fal.ai and Freepik, BFL aims to make Flux 1.1 Pro the AI ​​image generation model of choice for most users.

As BFL continues to push the boundaries of generative AI, the company is also expanding its workforce, seeking talented innovators to join its mission. Interested candidates can view vacancies via the company’s website.