### Pull Request Title
Introduce `max_output_tokens` Field for OpenAI Models
https://platform.deepseek.com/api-docs/news/news0725/#4-8k-max_tokens-betarelease-longer-possibilities
### Description
This commit introduces a new field `max_output_tokens` to the OpenAI
models, which allows specifying the maximum number of tokens that can be
generated in the output. This field is now integrated into the request
handling across multiple crates, ensuring that the output token limit is
respected during language model completions.
Changes include:
- Adding `max_output_tokens` to the `Custom` variant of the
`open_ai::Model` enum.
- Updating the `into_open_ai` method in `LanguageModelRequest` to accept
and use `max_output_tokens`.
- Modifying the `OpenAiLanguageModel` and `CloudLanguageModel`
implementations to pass `max_output_tokens` when converting requests.
- Ensuring that the `max_output_tokens` field is correctly serialized
and deserialized in relevant structures.
This enhancement provides more control over the output length of OpenAI
model responses, improving the flexibility and accuracy of language
model interactions.
### Changes
- Added `max_output_tokens` to the `Custom` variant of the
`open_ai::Model` enum.
- Updated the `into_open_ai` method in `LanguageModelRequest` to accept
and use `max_output_tokens`.
- Modified the `OpenAiLanguageModel` and `CloudLanguageModel`
implementations to pass `max_output_tokens` when converting requests.
- Ensured that the `max_output_tokens` field is correctly serialized and
deserialized in relevant structures.
### Related Issue
https://github.com/zed-industries/zed/pull/16358
### Screenshots / Media
N/A
### Checklist
- [x] Code compiles correctly.
- [x] All tests pass.
- [ ] Documentation has been updated accordingly.
- [ ] Additional tests have been added to cover new functionality.
- [ ] Relevant documentation has been updated or added.
### Release Notes
- Added `max_output_tokens` field to OpenAI models for controlling
output token length.
Release Notes:
- Adds support for Prompt Caching in Anthropic. For models that support
it this can dramatically lower cost while improving performance.
For future reference: WIP branch of copy/pasting a mixture of images and
text: https://github.com/zed-industries/zed/tree/copy-paste-images -
we'll come back to that one after landing this one.
Release Notes:
- You can now paste images into the Assistant Panel to include them as
context. Currently works only on Mac, and with Anthropic models. Future
support is planned for more models, operating systems, and image
clipboard operations.
---------
Co-authored-by: Antonio <antonio@zed.dev>
Co-authored-by: Mikayla <mikayla@zed.dev>
Co-authored-by: Jason <jason@zed.dev>
Co-authored-by: Kyle <kylek@zed.dev>
In this pull request, we change the zed.dev protocol so that we pass the
raw JSON for the specified provider directly to our server. This avoids
the need to define a protobuf message that's a superset of all these
formats.
@bennetbo: We also changed the settings for available_models under
zed.dev to be a flat format, because the nesting seemed too confusing.
Can you help us upgrade the local provider configuration to be
consistent with this? We do whatever we need to do when parsing the
settings to make this simple for users, even if it's a bit more complex
on our end. We want to use versioning to avoid breaking existing users,
but need to keep making progress.
```json
"zed.dev": {
"available_models": [
{
"provider": "anthropic",
"name": "some-newly-released-model-we-havent-added",
"max_tokens": 200000
}
]
}
```
Release Notes:
- N/A
---------
Co-authored-by: Nathan <nathan@zed.dev>
<img width="624" alt="image"
src="https://github.com/user-attachments/assets/f492b0bd-14c3-49e2-b2ff-dc78e52b0815">
- [x] Correctly set custom model token count
- [x] How to count tokens for Gemini models?
- [x] Feature flag zed.dev provider
- [x] Figure out how to configure custom models
- [ ] Update docs
Release Notes:
- Added support for quickly switching between multiple language model
providers in the assistant panel
---------
Co-authored-by: Antonio <antonio@zed.dev>
We will soon need `semantic_index` to be able to use
`CompletionProvider`. This is currently impossible due to a cyclic crate
dependency, because `CompletionProvider` lives in the `assistant` crate,
which depends on `semantic_index`.
This PR breaks the dependency cycle by extracting two crates out of
`assistant`: `language_model` and `completion`.
Only one piece of logic changed: [this
code](https://github.com/zed-industries/zed/commit/922fcaf5a6076e56890373035b1065b13512546d#diff-3857b3707687a4d585f1200eec4c34a7a079eae8d303b4ce5b4fce46234ace9fR61-R69).
* As of https://github.com/zed-industries/zed/pull/13276, whenever we
ask a given completion provider for its available models, OpenAI
providers would go and ask the global assistant settings whether the
user had configured an `available_models` setting, and if so, return
that.
* This PR changes it so that instead of eagerly asking the assistant
settings for this info (the new crate must not depend on `assistant`, or
else the dependency cycle would be back), OpenAI completion providers
now store the user-configured settings as part of their struct, and
whenever the settings change, we update the provider.
In theory, this change should not change user-visible behavior...but
since it's the only change in this large PR that's more than just moving
code around, I'm mentioning it here in case there's an unexpected
regression in practice! (cc @amtoaer in case you'd like to try out this
branch and verify that the feature is still working the way you expect.)
Release Notes:
- N/A
---------
Co-authored-by: Marshall Bowers <elliott.codes@gmail.com>
邻二氮杂菲
Roy Williams
Kyle Kelley
Roy Williams
Antonio ScandurraandNathan
Bennet Bo FennerandAntonio
