Black Forest Labs announced a $31M seed round and launched a suite of image-generation models

Black Forest Labs, an AI company based in Germany and founded by the creators of the technology that powers the original Stable Diffusion models, recently announced it secured $31 million in seed funding in a round led by Andreessen Horowitz, with the participation of several angel investors. The company also shared it received follow-up investments from General Catalyst and MätchVC.

In parallel with the funding announcement, Black Forest Labs also launched the FLUX.1 suite of text-to-image models. The FLUX.1 models feature detailed generations, prompt adherence, and scene complexity. The FLUX.1 suite includes three models: FLUX.1 [pro], a state-of-the-art model for commercial use, FLUX.1 [dev], an open-weight model for non-commercial applications distilled from FLUX.1 [pro] to achieve more efficiency while preserving quality and prompt adherence capabilities, and FLUX.1 [schnell] a small, openly available model covered by an Apache 2.0 license intended for local development and personal use.

The models feature a hybrid architecture of multimodal and parallel diffusion transformer blocks and a host of improvements that unlock state-of-the-art performance that surpasses popular competitors including Midjourney v6.0, DALL·E 3 (HD), and SD3-Ultra in aspects such as Visual Quality, Prompt Following, Size/Aspect Variability, Typography, and Output Diversity. The three models support several aspect ratios and resolutions from 0.1 to 2.0 megapixels.

Building on its commitment to revolutionizing the future of generative media, Black Forest Labs plans to use its text-to-image model suite as the foundation for a collection of state-of-the-art text-to-video generation systems that combine precision, high-definition editing, and speed.