The company is training large models with over 100 billion parameters to perform well on its internal benchmarks for reasoning. This training is being conducted on a compute cluster co-designed by Nvidia. Imbue is also investing in building its own AI and machine learning tooling, and conducting research into understanding the learning process in large language models. The company plans to use these tools and models to improve future, more general-purpose AI and to establish the groundwork for a platform that people can use to create their own custom models.
Key takeaways:
- Imbue, formerly known as Generally Intelligent, has raised $200 million in a Series B funding round, bringing its total raised to $220 million.
- The AI research lab is focusing on developing AI systems that can reason and code, with the aim of creating practical AI agents that can work in the real world.
- Imbue's AI models are being trained on data to reinforce good reasoning patterns and are optimized to perform well on internal benchmarks for reasoning.
- While not intending to productionize much of its current work, Imbue aims to use these tools and models to improve future, more general-purpose AI and establish a platform for people to create their own custom models.