tidely
tidely
When streaming using mlx_lm/server.py we should predict potential stop sequence matches, and generate tokens until we know that there is no match. This prevents the server from sending parts of...
The implementation of stop_criteria in mlx_lm.server is inherently flawed. Stop sequences only get matched when the newest tokens generated perfectly match a stop sequence. However it does not stop if...
### Check for existing issues - [X] Completed ### Describe the bug / provide steps to reproduce it Zed incorrectly auto indents `except` and `finally` blocks when the block contains...
We can combine platform specific requirements files by using Environment markers. These are specified in PEP 508. https://peps.python.org/pep-0508/#environment-markers
# Changelog Entry General code improvements Removed two time-of-check to time-of-use (TOCTOU) race conditions. ### Description The goal of this PR is to implement stricter type checks and add type...
# Pull Request Checklist **Before submitting, make sure you've checked the following:** - [x] **Target branch:** Please verify that the pull request targets the `dev` branch. - [x] **Description:** Provide...
The motivation behind this PR is to make more of the implicit logic explicit. In this case it means instead of treating 0 as a special case for the current...
The goal of this PR is to support tool calls using ollama. A lot of the serialization work was done in https://github.com/zed-industries/zed/pull/15803 however the abstraction over language models always disables...
Just a tiny patch to reduce allocations during context loading Calling `.cloned()` on an iterator clones each element one by one, while `into_iter().collect()` pre-allocates the resulting `Vec` Release Notes: -...
This PR moves the collab server from axum 0.6 to axum 0.8. The motivation behind this is to prevent buildup of technical debt by having years outdated dependencies. Axum has...