# language_models: Use weak entities in subscribe/observes around the language model registry \(\#52312\) · gitcafe/zed

[View on GitCafe](https://git.cafe/gitcafe/zed/commit/b28990e4719a46b4f0fd97590a2e8ca9acef4a5d)

Repository: [gitcafe/zed](https://git.cafe/gitcafe/zed)

Visibility: public

Requested revision: b28990e4719a46b4f0fd97590a2e8ca9acef4a5d

Requested commit: b28990e4719a46b4f0fd97590a2e8ca9acef4a5d

Commit: b28990e4719a46b4f0fd97590a2e8ca9acef4a5d

Tree: b51b1fe2ee6c2b4eb5b0352e265a9e90ad74c07f

Author: Ben Brandt

Committer: GitHub

## Message

```
language_models: Use weak entities in subscribe/observes around the language model registry (#52312)

## Context

I was getting some leak detection failures in evals and tracked it down
to these entities getting passed into observe/subscribe callbacks and
causing cycles.

Release Notes:

- N/A

Co-authored-by: Lukas Wirth <me@lukaswirth.dev>
```

## Parents

- [7c185202f5b32d45938107ccceadde395fdf1965](https://git.cafe/gitcafe/zed/commit/7c185202f5b32d45938107ccceadde395fdf1965?format=markdown)

[Source at this commit](https://git.cafe/gitcafe/zed/tree/b28990e4719a46b4f0fd97590a2e8ca9acef4a5d?format=markdown)