Startup Adaptive Raises $20M to Make Customizing Large Language Models Easier for Businesses
-
New startup Adaptive emerges from stealth with $20M in seed funding to make it easier for businesses to train customized large language models (LLMs).
-
Adaptive's platform improves on reinforcement learning from human feedback (RLHF) by capturing how real users interact with LLMs to tailor them.
-
The startup provides solutions to run reinforcement learning algorithms and control objectives, data, and algorithms - giving businesses more handle on AI.
-
Adaptive's tech works on top of open source or custom LLMs, helping test and monitor different models' performance.
-
The founding team previously worked on the popular open source Falcon LLMs and at Hugging Face before starting Adaptive.