AI Medical Agents
We make a number of different agents - each with a different approach - to experiment with. Each have a number of different running options. Try them here.
Try AI Agents →Medical Search
Here is a collection of search tools that the agents can use to lookup medical information. You can try them here.
Explore Search →Voice Chat
Voice will be important. Here we experiment with some voice prototypes - these are backed by the same agents, but translated to voice.
AI Evaluation Dashboard
Disclaimer: The evaluator is an LLM evaluator in beginning stages. Don't take these scores as gospel.
Open full evaluations Leaderboard in W&B →
Note: error counts are shown per cell. Weave errors reflect tracing failures; Case errors reflect our per-example errors.
Evaluation Version: v158
Loading evaluation stats...
Evaluation Comparison
Compare evaluation outputs side-by-side in a full-screen interface to understand how different agent configurations perform on the same inputs.
Open Comparison Tool →