| Recraft | |
|---|---|
 An image generated with Recraft V3. Prompt: cinematic shot, a colorful movie scene, neon lights, a man outside in dusk | |
| Developers | Recraft, Inc. |
| Initial release | May 31, 2023; 2 years ago (2023-05-31) |
| Stable release | V3 / October 30, 2024 (2024-10-30) |
| Operating system | Web-based |
| Type | Text-to-image model |
| Website | recraft |
Recraft is agenerative artificial intelligence program and service developed by the London-based startup Recraft, Inc. Like models such asMidjourney andDALL-E, it generates digital images from natural language prompts, but is specifically tailored for professional design workflows, with features that emphasize brand consistency, text fidelity, and layout control.[1]
Recraft, Inc. was founded in 2022 by machine learning scientist Anna Veronika Dorogush, best known for co-creating theCatBoost machine learning library atYandex.[2][3]
The company emerged from stealth on May 31, 2023, with a public release of its vector graphics generation capability onProduct Hunt. On January 17, 2024,TechCrunch profiled Recraft’s foundational model forgraphic design, noting its emphasis on addressing copyright and ethical concerns associated with AI-generated imagery.[1]
On October 28, 2024,TechCrunch reported that Recraft's third major model, V3, had topped a crowdsourced benchmark, surpassing Midjourney and OpenAI's DALL-E in overall image quality.[4]
On May 5, 2025, Recraft announced a $30 million Series B funding round led by Accel, reporting more than four million registered users at the time of the announcement.[5]
Recraft has not publicly disclosed the detailed technical architecture of its model. However, third-party reviews and benchmarks have noted that its performance resemblesdiffusion models such asStable Diffusion andMidjourney.[6][7]
The model is designed for creative workflows requiring visual consistency and flexible output formats. Reviewers have noted its ability to generate legible multi-line text,[7] produce high-resolution imagery at various canvas sizes,[8] and to maintain alignment with user-defined brand palettes and design themes.[5]
Though not open-source, Recraft offers its functionalities through a web interface and commercialAPI. Advanced features such as style settings and positioning control differentiate it from general-purposetext-to-image models.[5]
| Concepts |
| ||||||
|---|---|---|---|---|---|---|---|
| Applications | |||||||
| Implementations |
| ||||||
| People |
| ||||||
| Architectures |
| ||||||
