feat: Add Grok 4 Fast model to xAI provider #8476
Closed
+20
−0
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
- Added grok-4-fast model configuration to xaiModels in packages/types/src/providers/xai.ts - Added test case to verify grok-4-fast model selection works correctly - Model includes 2M context window and 30K max tokens support Fixes #8474
dosubot
bot
added
size:S
Author
|
Beginning a full review now - I will add inline comments shortly. |
2 tasks
![roomote[bot]](https://avatars.githubusercontent.com/in/1546624?s=60&v=4){kind=small}
This comment was marked as outdated.
This comment was marked as outdated.
Sorry, something went wrong.
hannesrudolph
added
the
Issue/PR - Triage
…xAI specs - Fix inputPrice from .0 to /bin/sh.20 per million tokens - Fix outputPrice from .0 to /bin/sh.50 per million tokens - Update maxTokens from 30,000 to 8,192 (matching grok-4, as xAI doesn't publish max output limit) - Remove subjective 'SOTA cost-efficiency' wording from description
Author
Review SummaryI found 1 critical issue that needs to be addressed:
|
| expect(model.id).toBe(testModelId) | ||
| expect(model.info).toEqual(xaiModels[testModelId]) | ||
| expect(model.info.contextWindow).toBe(2_000_000) | ||
| expect(model.info.maxTokens).toBe(30_000) |
Author
There was a problem hiding this comment.
Test expects maxTokens to be 30_000 but the model configuration (line 32 in xai.ts) defines it as 8192. This mismatch was introduced when the second commit updated the pricing and maxTokens but didn't update the test. The test will fail when run.
Suggested change
| expect(model.info.maxTokens).toBe(30_000) | |
| expect(model.id).toBe(testModelId) | |
| expect(model.info).toEqual(xaiModels[testModelId]) | |
| expect(model.info.contextWindow).toBe(2_000_000) | |
| expect(model.info.maxTokens).toBe(8192) |
The test was expecting 30,000 maxTokens but we corrected this to 8,192 to match grok-4, as xAI doesn't publish a maximum output token limit for grok-4-fast.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
This PR attempts to address Issue #8474 by adding support for the Grok 4 Fast model in the xAI provider.
Changes
grok-4-fastmodel configuration toxaiModelsinpackages/types/src/providers/xai.tsTesting
Related Issue
Fixes #8474
Notes
The model parameters (context window, max tokens, pricing) are based on the configuration already present in the Roo provider models. This enables users with xAI accounts to directly select and use Grok 4 Fast.
Feedback and guidance are welcome!
Important
Add
grok-4-fastmodel to xAI provider with specific parameters and test for correct model selection.grok-4-fastmodel toxaiModelsinxai.tswith 2,000,000 token context window, 8192 max tokens, and specific pricing.xai.spec.tsverifiesgrok-4-fastmodel selection.grok-4-fastmodel selection added and passes.This description was created by
for 7bea3b1. You can customize this summary. It will automatically update as commits are pushed.