Commit Graph
39 Commits
Author SHA1 Message Date
Peter KönigandBen Brandt 5dd8561b06 Fix DeepSeek Reasoner tool-call handling and add reasoning_content support (#44301)
## Closes #43887

## Release Notes:

### Problem
DeepSeek's reasoning mode API requires `reasoning_content` to be
included in assistant messages that precede tool calls. Without it, the
API returns a 400 error:

```
Missing `reasoning_content` field in the assistant message at message index 2
```

### Added/Fixed/Improved
- Add `reasoning_content` field to `RequestMessage::Assistant` in
`crates/deepseek/src/deepseek.rs`
- Accumulate thinking content from `MessageContent::Thinking` and attach
it to the next assistant/tool-call message
- Wire reasoning content through the language model provider in
`crates/language_models/src/provider/deepseek.rs`

### Testing
- Verified with DeepSeek Reasoner model using tool calls
- Confirmed reasoning content is properly included in API requests

Fixes tool-call errors when using DeepSeek's reasoning mode.

---------

Co-authored-by: Ben Brandt <benjamin.j.brandt@gmail.com>
2025-12-09 20:54:16 +01:00
Richard Feldman c0fadae881 Thought signatures (#42915)
Implement Gemini API's [thought
signatures](https://ai.google.dev/gemini-api/docs/thinking#signatures)

Release Notes:

- Added thought signatures for Gemini tool calls
2025-11-18 10:41:19 -05:00
Danilo Leal 2fb3d593bc agent_ui: Add component to standardize the configured LLM card (#42314)
This PR adds a new component to the `language_models` crate called
`ConfiguredApiCard`:

<img width="500" height="420" alt="Screenshot 2025-11-09 at 2  07@2x"
src="https://github.com/user-attachments/assets/655ea941-2df8-4489-a4da-bba34acf33a9"
/>

We were previously recreating this component from scratch with regular
divs in all LLM providers render function, which was redundant as they
all essentially looked the same and didn't have any major variations
aside from labels. We can clean up a bunch of similar code with this
change, which is cool!

Release Notes:

- N/A
2025-11-09 14:32:05 -03:00
chenmi cc1d66b530 agent_ui: Improve API key configuration UI display (#42306)
Improve the layout and text display of API key configuration in multiple
language model providers to ensure proper text wrapping and ellipsis
handling when API URLs are long.

Before:

<img width="320" alt="image"
src="https://github.com/user-attachments/assets/2f89182c-34a0-4f95-a43a-c2be98d34873"
/>

After:

<img width="320" alt="image"
src="https://github.com/user-attachments/assets/09bf5cc3-07f0-47bc-b21a-d84b8b1caa67"
/>

Changes include:
- Add proper flex layout with overflow handling
- Replace truncate_and_trailoff with CSS text ellipsis
- Ensure consistent UI behavior across all providers

Release Notes:

- Improved API key configuration display in language model settings
2025-11-09 13:00:31 -03:00
Danilo Leal 8f3da5c5cd settings_ui: Add pickers for theme and icon themes (#40829)
In the process of adding pickers for the theme and icon themes fields in
the settings UI, I felt like there was an improvement opportunity in
regards to where some of these components are stored. The `ui_input`
crate originally was meant only for the text field-like component, which
couldn't be in the regular `ui` crate due to the dependency with
`editor`. Given we had also added the number field there—which is
similar in also having the same dependency—it made sense to think of
this crate more like a home for form-like components rather than for
only one component.

However, we were also storing some settings UI-specific stuff in that
crate, which didn't feel right. So I ended up creating a new directory
within the `settings_ui` for components and moved all the pickers and
the custom input field there. I think this makes it for a cleaner
structure.

Release Notes:

- settings_ui: Added the ability to search for theme and icon themes in
their respective fields.
2025-10-21 19:58:43 -03:00
Michael Sloan 67984d5e49 provider configuration: Use SingleLineInput instead of Editor (#38814)
Release Notes:

- N/A
2025-09-25 22:38:27 +00:00
fcdab160f9 Settings refactor (#38367)
Co-Authored-By: Ben K <ben@zed.dev>
Co-Authored-By: Anthony <anthony@zed.dev>
Co-Authored-By: Mikayla <mikayla@zed.dev>

Release Notes:

- settings: Major internal changes to settings. The primary user-facing
effect is that some settings which did not make sense in project
settings files are no-longer read from there. (For example the inline
blame settings)

---------

Co-authored-by: Ben Kunkle <ben@zed.dev>
Co-authored-by: Mikayla Maki <mikayla.c.maki@gmail.com>
Co-authored-by: Anthony <anthony@zed.dev>
2025-09-18 16:47:23 +00:00
Michael Sloan a598fbaa73 ai: Show "API key configured for {URL}" for non-default urls (#38170)
Followup to #38163, also makes some changes intended to be included in
that PR.

Release Notes:

- N/A
2025-09-15 05:49:25 +00:00
Michael Sloan 634ae72cad Misc cleanup + clear language model provider API key editors when API keys are submitted (#38165)
Followup to #38163 along with some other misc cleanups

Release Notes:

- N/A
2025-09-15 05:08:38 +00:00
Michael Sloan 98edf1bf0b Reload API keys when URLs configured for LLM providers change (#38163)
Three motivations for this:

* Changing provider URL could cause credentials for the prior URL to be
sent to the new URL.
* The UI is in a misleading state after URL change - it shows a
configured API key, but on restart it will show no API key.
* #34110 will add support for both URL and key configuration for Ollama.
This is the first provider to have UI for setting the URL, and this
makes these issues show up more directly as odd UI interactions.

#37610 implemented something similar for the OpenAI and OpenAI
compatible providers. This extracts out some shared code, uses it in all
relevant providers, and adds more safety around key use.

I haven't tested all providers, but the per-provider changes were pretty
mechanical, so hopefully work properly.

Release Notes:

- Fixed handling of changes to LLM provider URL in settings to also load
the associated API key.
2025-09-15 03:36:24 +00:00
Bennet Bo FennerandAgus Zubiaga acb3406eb8 editor: Wrap placeholder if text overflows (#37919)
This fixes an issue where long placeholders would be cut off, e.g. in a
Claude Code thread:

<img width="387" height="115" alt="image"
src="https://github.com/user-attachments/assets/831a54aa-cf2b-4d87-af86-e368a5936f6b"
/>

Now:

<img width="354" height="115" alt="image"
src="https://github.com/user-attachments/assets/e5df5e05-0869-4db2-8dee-38611263191c"
/>


Most of the changes in this PR are caused by us requiring `&mut Window`
in `set_placeholder_text`.

Release Notes:

- Fixed an issue where placeholders inside editors would not wrap

---------

Co-authored-by: Agus Zubiaga <agus@zed.dev>
2025-09-10 15:38:19 +00:00
Piotr Osiewicz 9e0e233319 Fix clippy::needless_borrow lint violations (#36444)
Release Notes:

- N/A
2025-08-18 21:54:35 +00:00
Agus Zubiaga 8b89ea1a80 Handle auth for claude (#36442)
We'll now use the anthropic provider to get credentials for `claude` and
embed its configuration view in the panel when they are not present.

Release Notes:

- N/A
2025-08-18 20:40:59 +00:00
d497f52e17 agent: Improve error handling and retry for zed-provided models (#33565)
* Updates to `zed_llm_client-0.8.5` which adds support for `retry_after`
when anthropic provides it.

* Distinguishes upstream provider errors and rate limits from errors
that originate from zed's servers

* Moves `LanguageModelCompletionError::BadInputJson` to
`LanguageModelCompletionEvent::ToolUseJsonParseError`. While arguably
this is an error case, the logic in thread is cleaner with this move.
There is also precedent for inclusion of errors in the event type -
`CompletionRequestStatus::Failed` is how cloud errors arrive.

* Updates `PROVIDER_ID` / `PROVIDER_NAME` constants to use proper types
instead of `&str`, since they can be constructed in a const fashion.

* Removes use of `CLIENT_SUPPORTS_EXA_WEB_SEARCH_PROVIDER_HEADER_NAME`
as the server no longer reads this header and just defaults to that
behavior.

Release notes for this is covered by #33275

Release Notes:

- N/A

---------

Co-authored-by: Richard Feldman <oss@rtfeldman.com>
Co-authored-by: Richard <richard@zed.dev>
2025-06-30 21:01:32 -06:00
Umesh Yadav 108162423d language_models: Emit UsageUpdate events for token usage in DeepSeek and OpenAI (#33242)
Closes #ISSUE

Release Notes:

- N/A
2025-06-25 09:42:30 +02:00
Richard Feldman 5405c2c2d3 Standardize on u64 for token counts (#32869)
Previously we were using a mix of `u32` and `usize`, e.g. `max_tokens:
usize, max_output_tokens: Option<u32>` in the same `struct`.

Although [tiktoken](https://github.com/openai/tiktoken) uses `usize`,
token counts should be consistent across targets (e.g. the same model
doesn't suddenly get a smaller context window if you're compiling for
wasm32), and these token counts could end up getting serialized using a
binary protocol, so `usize` is not the right choice for token counts.

I chose to standardize on `u64` over `u32` because we don't store many
of them (so the extra size should be insignificant) and future models
may exceed `u32::MAX` tokens.

Release Notes:

- N/A
2025-06-17 10:43:07 -04:00
Ben Brandt e4bd115a63 More resilient eval (#32257)
Bubbles up rate limit information so that we can retry after a certain
duration if needed higher up in the stack.

Also caps the number of concurrent evals running at once to also help.

Release Notes:

- N/A
2025-06-09 18:07:22 +00:00
Umesh Yadav c75ad2fd11 language_models: Add thinking support to DeepSeek provider (#32338)
For DeepSeek provider thinking is returned as reasoning_content and we
don't have to send the reasoning_content back in the request.

Release Notes:

- Add thinking support to DeepSeek provider
2025-06-09 11:10:55 +02:00
THELOSTSOULandBennet Bo Fenner b820aa1fcd Add tool support for DeepSeek (#30223)
[deepseek function call
api](https://api-docs.deepseek.com/guides/function_calling)
has been released and it is same as openai.

Release Notes:

- Added tool calling support for Deepseek Models

---------

Co-authored-by: Bennet Bo Fenner <bennetbo@gmx.de>
2025-06-03 10:59:36 +02:00
Kirill Bulatov 16366cf9f2 Use anyhow more idiomatically (#31052)
https://github.com/zed-industries/zed/issues/30972 brought up another
case where our context is not enough to track the actual source of the
issue: we get a general top-level error without inner error.

The reason for this was `.ok_or_else(|| anyhow!("failed to read HEAD
SHA"))?; ` on the top level.

The PR finally reworks the way we use anyhow to reduce such issues (or
at least make it simpler to bubble them up later in a fix).
On top of that, uses a few more anyhow methods for better readability.

* `.ok_or_else(|| anyhow!("..."))`, `map_err` and other similar error
conversion/option reporting cases are replaced with `context` and
`with_context` calls
* in addition to that, various `anyhow!("failed to do ...")` are
stripped with `.context("Doing ...")` messages instead to remove the
parasitic `failed to` text
* `anyhow::ensure!` is used instead of `if ... { return Err(...); }`
calls
* `anyhow::bail!` is used instead of `return Err(anyhow!(...));`

Release Notes:

- N/A
2025-05-20 23:06:07 +00:00
8fdf309a4a Have read_file support images (#30435)
This is very basic support for them. There are a number of other TODOs
before this is really a first-class supported feature, so not adding any
release notes for it; for now, this PR just makes it so that if
read_file tries to read a PNG (which has come up in practice), it at
least correctly sends it to Anthropic instead of messing up.

This also lays the groundwork for future PRs for more first-class
support for images in tool calls across more image file formats and LLM
providers.

Release Notes:

- N/A

---------

Co-authored-by: Agus Zubiaga <hi@aguz.me>
Co-authored-by: Agus Zubiaga <agus@zed.dev>
2025-05-13 10:58:00 +02:00
Antonio Scandurra 9f6809a28d Reuse conversation cache when streaming edits (#30245)
Release Notes:

- Improved latency when the agent applies edits.
2025-05-08 14:36:34 +02:00
720dfee803 Treat invalid JSON in tool calls as failed tool calls (#29375)
Release Notes:

- N/A

---------

Co-authored-by: Max <max@zed.dev>
Co-authored-by: Max Brunsfeld <maxbrunsfeld@gmail.com>
2025-04-24 16:54:27 -04:00
Michael Sloan fbf7caf93e Default to fast model for thread summaries and titles + don't include system prompt / context / thinking segments (#29102)
* Adds a fast / cheaper model to providers and defaults thread
summarization to this model. Initial motivation for this was that
https://github.com/zed-industries/zed/pull/29099 would cause these
requests to fail when used with a thinking model. It doesn't seem
correct to use a thinking model for summarization.

* Skips system prompt, context, and thinking segments.

* If tool use is happening, allows 2 tool uses + one more agent response
before summarizing.

Downside of this is that there was potential for some prefix cache reuse
before, especially for title summarization (thread summarization omitted
tool results and so would not share a prefix for those). This seems fine
as these requests should typically be fairly small. Even for full thread
summarization, skipping all tool use / context should greatly reduce the
token use.

Release Notes:

- N/A
2025-04-19 23:26:29 +00:00
Danilo Leal e27f6a984f agent: Simplify design of the settings view (#29041)
Containing everything in boxes wasn't super necessary here. Want to
still improve the switch color contrast here, but will probably do that
in a separate PR.

<img
src="https://github.com/user-attachments/assets/f826a7a8-beaf-45d0-9dc2-36dc210c418e"
width="700"/>

Release Notes:

- N/A
2025-04-18 14:24:53 -03:00
Marshall Bowers 889bc13b7d language_model: Remove use_any_tool method from LanguageModel (#27930)
This PR removes the `use_any_tool` method from the `LanguageModel`
trait.

It was not being used anywhere, and doesn't really fit in our new tool
use story.

Release Notes:

- N/A
2025-04-02 15:49:21 +00:00
Marshall Bowers 5880271b11 language_model: Add supports_tools method to LanguageModel (#27867)
This PR adds a new `supports_tools` method to the `LanguageModel` trait
to indicate whether a given model supports tool use.

Release Notes:

- N/A
2025-04-01 19:56:05 +00:00
Piotr Osiewicz dc64ec9cc8 chore: Bump Rust edition to 2024 (#27800)
Follow-up to https://github.com/zed-industries/zed/pull/27791

Release Notes:

- N/A
2025-03-31 20:55:27 +02:00
Mikayla Maki 1aefa5178b Move "async move" a few characters to the left in cx.spawn() (#26758)
This is the core change:
https://github.com/zed-industries/zed/pull/26758/files#diff-044302c0d57147af17e68a0009fee3e8dcdfb4f32c27a915e70cfa80e987f765R1052

TODO:
- [x] Use AsyncFn instead of Fn() -> Future in GPUI spawn methods
- [x] Implement it in the whole app
- [x] Implement it in the debugger 
- [x] Glance at the RPC crate, and see if those box future methods can
be switched over. Answer: It can't directly, as you can't make an
AsyncFn* into a trait object. There's ways around that, but they're all
more complex than just keeping the code as is.
- [ ] Fix platform specific code

Release Notes:

- N/A
2025-03-19 02:09:02 +00:00
Marshall Bowers aceab76ae4 gpui: Rename rounded_md to rounded_sm (#26228)
This PR renames the `rounded_md` style method to `rounded_sm`.

Follow up to https://github.com/zed-industries/zed/pull/26221, which
freed up the `rounded_sm` name.

Release Notes:

- N/A
2025-03-06 17:57:31 +00:00
Danilo Leal 508b581215 assistant: Refine settings view's instruction visuals (#25812)
I've been bothered by using simple hyphens for bullet lists here for a
while; it kinda looked cheap and not well-formatted. So, in this PR, I'm
adding a new, custom UI component in the `language_models` crate, called
`InstructionListItem`, based off the `ListItem` that's somewhat
mimic'ing what a `<li>` would be on the web.

It does have a "rigid" structure as in it's always a label followed by a
button (which is optional), but that seems okay given it has been the
overall shape of the copy we've been using here. Also, never really
loved that we were pasting URLs directly, that kinda felt cheap, too. I
could see an argument where it's just clearer, but it looks too
cluttered, as URLs aren't super pretty, necessarily.

| Before | After |
|--------|--------|
| <img
src="https://github.com/user-attachments/assets/ffd1ac27-b1f4-450d-abf5-079285fc9877"
width="700px" /> | <img
src="https://github.com/user-attachments/assets/28fb9d0d-205d-45d8-9e43-1aaa947adc96"
width="700px" /> |

Release Notes:

- N/A
2025-02-28 12:06:47 -03:00
Marshall Bowers 0acd556106 language_model: Remove dependencies on individual model provider crates (#25503)
This PR removes the dependencies on the individual model provider crates
from the `language_model` crate.

The various conversion methods for converting a `LanguageModelRequest`
into its provider-specific request type have been inlined into the
various provider modules in the `language_models` crate.

The model providers we provide via Zed's cloud offering get to stay, for
now.

Release Notes:

- N/A
2025-02-24 16:41:35 -05:00
Antonio Scandurra f517050548 Partially fix assistant onboarding (#25313)
While investigating #24896, I noticed two issues:

1. The default configuration for the `zed.dev` provider was using the
wrong string for Claude 3.5 Sonnet. This meant the provider would always
result as not configured until the user selected it from the model
picker, because we couldn't deserialize that string to a valid
`anthropic::Model` enum variant.
2. When clicking on `Open New Chat`/`Start New Thread` in the provider
configuration, we would select `Claude 3.5 Haiku` by default instead of
Claude 3.5 Sonnet.

Release Notes:

- Fixed some issues that caused AI providers to sometimes be
misconfigured.
2025-02-24 07:29:55 +00:00
Marshall Bowers 21bb7242ea Add CredentialsProvider to silence keychain prompts in development (#25266)
This PR adds a new `CredentialsProvider` trait that abstracts over
interacting with the system keychain.

We had previously introduced a version of this scoped just to Zed auth
in https://github.com/zed-industries/zed/pull/11505.

However, after landing https://github.com/zed-industries/zed/pull/25123,
we now have a similar issue with the credentials for language model
providers that are also stored in the keychain (and thus also produce a
spam of popups when running a development build of Zed).

This PR takes the existing approach and makes it more generic, such that
we can use it everywhere that we need to read/store credentials in the
keychain.

There are still two credential provider implementations:

- `KeychainCredentialsProvider` will interact with the system keychain
(using the existing GPUI APIs)
- `DevelopmentCredentialsProvider` will use a local file on the file
system

We only use the `DevelopmentCredentialsProvider` when:

1. We are running a development build of Zed
2. The `ZED_DEVELOPMENT_AUTH` environment variable is set
- I am considering removing the need for this and making it the default,
but that will be explored in a follow-up PR.

Release Notes:

- N/A
2025-02-20 17:58:50 +00:00
Marshall Bowers 7a6b652ebc language_model: Return AuthenticateErrors from LanguageModelProvider::authenticate (#25126)
This PR updates the `LanguageModelProvider::authenticate` method to
return an `AuthenticateError` instead of an `anyhow::Error`.

This allows us to model the "credentials not found" state explicitly as
`AuthenticateError::CredentialsNotFound`, which enables the caller to
check for this state and act accordingly.

Planning to use this in #25123 to silence errors about missing
credentials when authenticating providers in the background.

Release Notes:

- N/A
2025-02-19 00:01:48 +00:00
Michael Sloan b1872e3afd cx.background_executor().spawn(...) -> cx.background_spawn(...) (#25103)
Done automatically with

> ast-grep -p '$A.background_executor().spawn($B)' -r
'$A.background_spawn($B)' --update-all --globs "\!crates/gpui"

Followed by:

* `cargo fmt`
* Unexpected need to remove some trailing whitespace.
* Manually adding imports of `gpui::{AppContext as _}` which provides
`background_spawn`
* Added `AppContext as _` to existing use of `AppContext`

Release Notes:

- N/A
2025-02-18 20:30:33 +00:00
Jason Lee 706f7be5e7 gpui: Add line_clamp to truncate text after a specified number of lines (#23058)
Release Notes:

- N/A

Add this feature for some case we need keep 2 or 3 lines, but truncate.
For example the blog post summary.

- Added `line_clamp` method.
    Ref: https://tailwindcss.com/docs/line-clamp


## Break changes:

- Renamed `gpui::Truncate` to `gpui::TextOverflow` to match
[CSS](https://developer.mozilla.org/en-US/docs/Web/CSS/text-overflow).
- Update `truncate` style method to match [Tailwind
CSS](https://tailwindcss.com/docs/text-overflow) behavior:

    ```css
    overflow: hidden;
    text-overflow: ellipsis;
    white-space: nowrap;
    ```
<img width="538" alt="image"
src="https://github.com/user-attachments/assets/c69c4213-eac9-4087-9daa-ce7afe18c758"
/>


## Show case

<img width="816" alt="image"
src="https://github.com/user-attachments/assets/e0660290-8042-4954-b93c-c729d609484a"
/>

![CleanShot 2025-01-13 at 17 22
05](https://github.com/user-attachments/assets/38644892-79fe-4254-af9e-88c1349561bd)

## Describe changes

The [second
commit](https://github.com/zed-industries/zed/commit/6b41c2772f9d1c5f4efe69fe83206a14407af4d4)
for make sure text layout to match with the line clamp. Before this
change, they may wrap many lines in sometimes. And I also make
line_clamp default to 1 if we used `truncate` to ensure no wrap.

> TODO: There is still a tiny detail that is not easy to fix. This
problem only occurs in the case of certain long words. I will think
about how to improve it later. At present, this has some flaws but does
not affect the use.
2025-01-29 22:14:24 +02:00
Danilo Leal f8dddf0a5c assistant2: Tweak the settings UI (#23845)
This PR does some somewhat light UI adjustment to the Assistant 2
settings view. The Prompt Library section should feature the default
prompts in the future, so that's why it's been separated that way.

<img width="800" alt="Screenshot 2025-01-29 at 2 59 59 PM"
src="https://github.com/user-attachments/assets/7b033bde-51ab-44d5-9e53-3f72b8ff5f51"
/>

Release Notes:

- N/A
2025-01-29 16:20:09 -03:00
邻二氮杂菲andMarshall Bowers 29bfb56739 Add DeepSeek support (#23551)
- Added support for DeepSeek as a new language model provider in Zed
Assistant
- Implemented streaming API support for real-time responses from
DeepSeek models.
- Added a configuration UI for DeepSeek API key management and settings.
- Updated documentation with detailed setup instructions for DeepSeek
integration.
- Added DeepSeek-specific icons and model definitions for seamless
integration into the Zed UI.
- Integrated DeepSeek into the language model registry, making it
available alongside other providers like OpenAI and Anthropic.

Release Notes:

- Added support for DeepSeek to the Assistant.

---------

Co-authored-by: Marshall Bowers <git@maxdeviant.com>
2025-01-27 13:40:59 -05:00