Samuel Colvin

icon indicating a website link https://pydantic.dev/logfire icon indicating an email [email protected]

icon company @pydantic icon location London, United Kingdom icon location The Pydantic Stack: Pydantic Validation, Pydantic AI and Pydantic Logfire

Results 1283 comments of Samuel Colvin

V3 - act like protocol buffer

no, we'd need to add it, but if we added it to Jiter correctly, it might require minimal extra logic in pydantic-core.

Mistral Support

@YanSte `BaseModel` isn't yet supported by partial validation, this is explained in data at https://docs.pydantic.dev/latest/concepts/experimental/#typeadapter-only. Use a `TypedDict` and it should work.

Mistral Support

Thanks so much!

An assistant message with 'tool_calls' must be followed by tool messages

I think this might be fixed in the latest pydantic-ai, what version are you using? If it's still occurring with the latest version, I would need an MRE to be...

uv.lock has pinned pyscopg-binary==3.2.3, which can't be build on Mac M1

we definitely want the `uv.lock` file committed!

Format the console log

Sorry for the slow reply, I'll reply tomorrow.

Format the console log

@NourEldin-Osama I think you're right that we don't support this at the moment. I'm sure we can fix logfire so loguru continues to behave as it would if logfire wasn't...

How can one simply return the response of the tool, instead of routing the response to a final result handler?

I don't really understand the question I'm afraid. Result tools are not required - if the return type is `str`, no tool is used. And there aren't any LLM calls...

Fix macos-latest and 3.7 tests

Merged, but we should drop 3.7 and 3.8 anyway.

Incorrect Generated JSON Schema for model with custom Dict

This would all be a lot easier to treat if the code snippets were highlighted. I'm really not clear what the issue is, please can you consolidate into an MRE.