pydantic.dev

AIAI-generated summary Pydantic AI explains the differences between code mode and sandboxes for AI agents. The post details how workspaces in Pydantic AI allow moving boundaries and running code in Monty.

AIAI-generated summary Pydantic states its open-source Pydantic AI Harness provides over eighty capabilities across Pydantic AI and the harness package for Python agents. Pydantic says the harness includes file and shell tools, planning, memory, sandboxes, and remote workspaces supporting multiple model providers and execution engines.

AIAI-generated summary Pydantic released a 5,276-run benchmark comparing code mode and traditional tool calling for AI agents. Pydantic says the results show that models perform better when writing loops rather than acting as the loop themselves.