short circuit
@un3valuated
agentic systems & python engineer
456 Following 42 Followers
LangChain's adapter with pass@1 scored almost same on both accuracy & cost but spent 1 hour and half more time.
Terminal-Bench 2.1 pass@1 run is scored 76.40% with Luna High using Pydantic AI adapter of my Agent SDK.
Next one will have same experiment with LangChain adapter and we'll see the difference.
Show more
@samuelcolvin @pydantic Lately I'm powering my filesystem simulator VSH with Monty ( to track the side effects of filesystem operations and review changes before running them (using Monty's filesystem mounts), and currently working on its Pydantic AI capability.
Show more
