Cambridge spinout Flower Labs launches AI model to rival OpenAI, after £23M raised
What's the deal? Flower Labs, a Cambridge University spinout, has launched Endeavor 1.0, a "frontier-class" generalist AI model it says competes with the leading systems from OpenAI and AnthropicDealroom has a profile for this one. Try Dealroom →. The London and Hamburg-based startup, founded in 2023, has raised over £23 million and lets companies train AI models on data that stays in place rather than moving to a central database.
What's the product? Endeavor is built on open-source models, combined with Flower Labs' proprietary reasoning model and in-house software. It can be deployed on a company's own IT systems — a hospital, for example, could keep patient data within its own network — giving organisations greater control and security.
Why now? The launch lands amid a UK government push to back domestic alternatives to Silicon Valley AI giants, and a wider European drive to deploy AI locally with tighter data control. "Europe should not have to rent its intelligence indefinitely from a handful of US companies," co-founder and chief scientist Nicholas Lane, a Cambridge machine learning professor, told the Times.
What's the endgame? Flower Labs, whose clients include the NHSDealroom has a profile for this one. Try Dealroom → and JP Morgan, bills itself as a full-stack AI startup building open-source models, agents, and infrastructure. It says Endeavor matches OpenAI's GPT-5.6 Sol and Anthropic's Claude Fable 5 on some tasks, and outperforms Moonshot's Kimi K3 on some benchmarks.
What's the rollout? The model is launching first to a select group of organisations, ahead of a broader release as the startup ramps up its compute. It will be available under licence, with clients able to train it on data that stays out of any centralised database.
The signal: Flower Labs is pitching Endeavor as a sovereign alternative to US closed-source models accessed via an API, tapping European demand for AI that firms can own rather than rent. Its founders, Y Combinator alumni, are betting that local deployment and data control will matter more as governments push for independence from a handful of American providers.
Read more: Tech.eu
Image credit: Cory M. Grenier
