F13

Models that turn text prompts into editable SVG charts, infographics, diagrams and on-brand documents in seconds, grounded in real data.

F13: Waitlist, invite only. Get it at f13.com.

F13 is a Berlin-based foundational lab training models that natively generate vector graphics, aimed at teams that need accurate, production-grade visuals. Users describe what they need and receive a clean, editable SVG in seconds. Outputs are grounded in data supplied by the user and in real-world information, and feeding in a design system returns on-brand material. Initial capabilities cover data visualization and editorial infographics, with diagrams, geospatial visuals and brand-consistent documents built on the same foundation. The models can be embedded in AI assistants, documents and products; the first model's public release is upcoming via a waitlist.