tidely

Is there anything against moving the logging setup before the torch import? Currently when the torch import fails we save the error in a global variable, which isn't cleaned up...

https://github.com/zed-industries/zed/issues/10832

The best way to allow tool calls on specific models is likely to have the user explicitly allow tools on specific models in `settings.json` under the ollama provider. However this...

> The [show model information API](https://github.com/ollama/ollama/blob/main/docs/api.md#show-model-information) should have what you're looking for. You'll just have to call it on each model. Unfortunately I believe the `show` endpoint does not contain...

I just checked with a bunch of different model families are it seems I was wrong! Thank you @lj3954, I'll work on something to make use of it.

@mgsloan Hey! I saw you recently worked on https://github.com/zed-industries/zed/pull/29885. I looked at creating a similar map_to_completion_events function for the ollama provider, but it doesn't seem to be working as I...

I think having additional indirection in examples complicates the learning process. I think most people want to learn incrementally and having an additional 50-100 lines of tests on examples equivalent...

Could be made `const` compatible by avoiding `.try_into()` and converting `&[u8]` into `[u8; 4]` manually. We actually do hardcode some tags, which would benefit from `const`. I think this more...

Tests failing is unrelated

There's a rather annoying quirk where if you define a custom model, but don't provide a value for `supports_tools`, it will default to false, and override what ollama returns. This...