Closes#36341
<img width="543" height="548" alt="image"
src="https://github.com/user-attachments/assets/ab76a32c-c622-4025-9b28-5accc8d3f04c"
/>
In the case where commit message was suggested based on single tracked
entry, this PR adds a clause to the condition to ensure there are no
staged entries.
Release Notes:
- Fixed commit message suggestion when there is one unstaged tracked
file, but multiple untracked files are staged.
- **WIP: reorganize dispositions**
- **Introduce a LocalToolchainStore trait and use it for LspAdapter
methods**
Closes#35782Closes#27331
Release Notes:
- Python: Improved propagation of a selected virtual environment into
the LSP configuration. This should the make all language-related
features such as Go to definition or Find all references more reliable.
---------
Co-authored-by: Cole Miller <cole@zed.dev>
Co-authored-by: Lukas Wirth <lukas@zed.dev>
### TL;DR
* Adds `capabilities` configuration for OpenAI-compatible models
* Relates to
https://github.com/zed-industries/zed/issues/36215#issuecomment-3193920491
### Summary
This PR introduces support for configuring model capabilities for
OpenAI-compatible language models. The implementation addresses the
issue that not all OpenAI-compatible APIs support the same features -
for example, Cerebras' API explicitly does not support
`parallel_tool_calls` as documented in their [OpenAI compatibility
guide](https://inference-docs.cerebras.ai/resources/openai#currently-unsupported-openai-features).
### Changes
1. **Model Capabilities Structure**:
- Added `ModelCapabilityToggles` struct for UI representation with
boolean toggle states
- Implemented proper parsing of capability toggles into
`ModelCapabilities`
2. **UI Updates**:
- Modified the "Add LLM Provider" modal to include checkboxes for each
capability
- Each OpenAI-compatible model can now be configured with its specific
capabilities through the UI
3. **Configuration File Structure**:
- Updated the settings schema to support a `capabilities` object for
each `openai_compatible` model
- Each capability (`tools`, `images`, `parallel_tool_calls`,
`prompt_cache_key`) can be individually specified per model
### Example Configuration
```json
{
"openai_compatible": {
"Cerebras": {
"api_url": "https://api.cerebras.ai/v1",
"available_models": [
{
"name": "gpt-oss-120b",
"max_tokens": 131000,
"capabilities": {
"tools": true,
"images": false,
"parallel_tool_calls": false,
"prompt_cache_key": false
}
}
]
}
}
}
```
### Tests Added
- Added tests to verify default capability values are correctly applied
- Added tests to verify that deselected toggles are properly parsed as
`false`
- Added tests to verify that mixed capability selections work correctly
Thanks to @osyvokon for the desired `capabilities` configuration
structure!
Release Notes:
- OpenAI-compatible models now have configurable capabilities (#36370;
thanks @calesennett)
---------
Co-authored-by: Oleksiy Syvokon <oleksiy@zed.dev>
Closes#36229
Fix zero-sized texture creation that triggers a SIGABRT in the Metal
renderer. Not sure why this happens yet, but it likely occurs when
`native_window.contentView()` returns a zero `NSSize` during initial
window creation, before the view size is computed.
Release Notes:
- Fixed a rare startup crash on macOS.
This PR adds the git clone action to the project panel. It also changes
the "open" button to open a folder instead of the recent projects modal,
which feels faster to start with, more intuitive, and also consistent
with VS Code (which I think is good in this specific case).
<img width="500" height="1334" alt="CleanShot 2025-08-17 at 2 10 01@2x"
src="https://github.com/user-attachments/assets/ff953228-9e8e-413b-89ba-fa0870a0df17"
/>
Release Notes:
- Improved the project panel empty state by including the git clone
action and allowing users to quickly open a local folder.
After redesigning all Zed icons
(https://github.com/zed-industries/zed/pull/35856), it felt like using
1.5 for stroke width didn't really flow well with the default typeface
default font weight. Reducing it to 1.2 makes the UI much sharper, less
burry, and more cohesive overall.
Release Notes:
- N/A
Closes#36097
Release Notes:
- Fixed API key input fields getting shrunk in Agent Panel settings view
on low panel widths paired with high UI font sizes.
This PR removes the `/docs` slash command.
We never fully shipped this—with it requiring explicit opt-in via a
setting—and it doesn't seem like the feature is needed in an agentic
world.
Release Notes:
- Removed the `/docs` slash command.
Ensures issues like #36242 and #36295 do not arise where users are
confused that the agent panel does not follow the default UI font size
whilst also keeping the possibility of customization. The agent font
size was matching the UI font size previously alredy, which makes it
easier to change it for most scenarios.
Also cleans up some related logic around modifying the font sizes.
Release Notes:
- The agent panel font size will now inherit the UI font size by default
if not set in your settings.
This PR removes the billing-related tables from the SQLite schema, as we
don't actually reference these tables anywhere in the Collab codebase
anymore.
Release Notes:
- N/A
This PR updates the `admin` column on the `users` table to be
non-nullable.
We were already treating it like this in practice.
All rows in the production database already have a value for the `admin`
column.
Release Notes:
- N/A
Closes#36297
While we set the editor as read-only for bundled files, we didn't do
this for the underlying buffer. This PR fixes this and adds a test for
the corresponding case.
Release Notes:
- Fixed an issue where bundled files (e.g. the default settings) could
be edited in some circumstances
The minidump-based crash reporting is now entirely separate from our
legacy panic_hook-based reporting. This should improve the association
of minidumps with their metadata and give us more consistent crash
reports.
Release Notes:
- N/A
---------
Co-authored-by: Max Brunsfeld <maxbrunsfeld@gmail.com>
Release Notes:
- Added an option for the status_bar.cursor_position_button. Setting to
`false` will hide the button. It defaults to `true`.
This builds off the recent work to hide the language selection button
(https://github.com/zed-industries/zed/pull/33977). I tried to follow
that pattern, and to pick a clear name for the option, but any
feedback/change is welcome.
---------
Co-authored-by: zumbalogy <3770982+zumbalogy@users.noreply.github.com>
Closes #ISSUE
Release Notes:
- Keymap Editor: Fixed an issue where leaving the arguments field empty
would result in an error even if arguments were optional
Follow-up to https://github.com/zed-industries/zed/pull/36233
The above PR simplified the handling but introduced some bugs: The
replace buttons were no longer clickable, some buttons also lost their
toggle states, some buttons shared their element id and, lastly, some
buttons were clickable but would not trigger the right action. This PR
fixes all that.
Release Notes:
- N/A
This PR increases the minimum required version to connect to Collab.
Previously this was set at v0.157.0.
The new minimum required version is v0.198.4, which is the first version
where we no longer connect to Collab automatically.
Clients on the v0.199.x minor version will also need to be v0.199.2 or
greater in order to connect, due to us hotfixing the connection changes
to the Preview branch.
We're doing this to force clients to upgrade in order to connect to
Collab, as we're going to be removing some of the old RPC usages related
to authentication that are no longer used. Therefore, we want users to
be on a version of Zed that does not rely on those messages.
Users will see a message similar to this one, prompting them to upgrade:
<img width="1209" height="875" alt="Screenshot 2025-08-15 at 11 37
55 AM"
src="https://github.com/user-attachments/assets/59ffff3e-8f82-4152-84a8-776c691eaaee"
/>
> Note: In this case I'm simulating the error state, which is why I'm
signed in via Cloud while still not being able to connect to Collab.
Users on older versions will see the "Please update Zed to Collaborate"
message without being signed in.
Release Notes:
- N/A
1. Previously, checkpoints only appeared when an agent's edit happened
immediately after a user message. This is rare (agent usually collects
some context first), so they were almost never shown. This is now fixed.
2. After this change, a checkpoint is created after every edit
operation. So when the agent edits files five times in a single dialog
turn, we will now display five checkpoints.
As a bonus, it's now possible to undo only a part of a long agent
response.
Closes#36092, #32917
Release Notes:
- Create agent checkpoints more frequently (before every edit)
