-
Notifications
You must be signed in to change notification settings - Fork 6.9k
Baseten provider #5238
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Baseten provider #5238
Conversation
🦋 Changeset detectedLatest commit: 76f93e8 The changes in this PR will be included in the next version bump. This PR includes changesets to release 1 package
Not sure what this means? Click here to learn what changesets are. Click here if you're a maintainer who wants to add another changeset to this PR |
Hey! Thanks for working on this @AlexKer! This is directionally correct, but looks like you got avalanched with merge conflicts. Do you mind resolving those and ping me for more thorough review + testing? |
@pashpashpash looks like a new provider |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Fix the token counting logic but emitting usage once
@arafatkatze @pashpashpash fixed! Thanks for your review, please merge when ready! |
@AlexKer Your API is not persisted on the frontend. Please test the PR and fix the frontend and make sure you cna actually make a request that goes through |
@arafatkatze fixed, also added support for 2 new qwen3 models we added the past week. Here is a video showing it working: https://www.loom.com/share/ba73fcbe26e6477490852676bb096527?t=11&sid=2167adb0-880f-4b3e-beef-5ada27378e91 |
@arafatkatze thanks for catching those! changed. |
@AlexKer This is a fast provider and I liked playing with it. I did not verify that the maxTokens, contextWindow and pricing of the models is accurate based on your documentation. I can merge this PR provided you take the responsibility of their accuracy. |
@arafatkatze I'm glad you enjoyed testing it -- much appreciated for your fast feedback. I can take responsibility in the accuracy based on our docs. We will look into a way of dynamically updating these values in the future. |
Merged |
This PR adds Baseten Model APIs with support of the DeepSeek, Llama and Kimi K2 models. We can now select Baseten as a provider in the Cline dropdown menu. This is similar to #4943.
This was tested locally with BASETEN_API_KEY and various unit tests (not included in PR).
Pre-flight Checklist
npm test
) and code is formatted and linted (npm run format && npm run lint
)npm run changeset
(required for user-facing changes)Screenshots
Loom demo link: https://www.loom.com/share/acef33c66eca4a9494e947924fda79f0
Additional Notes
Important
Adds Baseten as a new provider with model support and integrates it into the API, state management, and UI components.
ApiOptions.tsx
andBasetenProvider.tsx
.baseten.ts
.refreshBasetenModels.ts
.models.proto
andapi.ts
.api-configuration-conversion.ts
for Baseten model info conversion.state.ts
andstate-keys.ts
to handle Baseten API keys and model configurations.ExtensionStateContext.tsx
to manage Baseten models.BasetenModelPicker.tsx
for model selection.providerUtils.ts
for Baseten model normalization.README.md
to include Baseten in supported providers list.This description was created by
for f34c120. You can customize this summary. It will automatically update as commits are pushed.