fix: calibrate input token when using anthropic models of sap ai core… #5469
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
🦋 Changeset detectedLatest commit: 419953b 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 |
7148f5e to
419953b
Compare
saoudrizwan
approved these changes
Aug 9, 2025
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Outdated suggestions cannot be applied.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.
Related Issue
Issue: #XXXX
No issue is created yet, though there are discussions about token count information issue in PR #4683 and #5399.
Here is a screenshot to show that after prompt cache is enabled for anthropic model, the context usage is much fewer than the real situation:
As it shows, the input token is 4, and the context window usage is 4 + 1.0k ≈ 1.0K, which is not correct as the size of cline system prompt is more than 4 tokens.
Description
After #5399 is merged, there is a token counting issue
The root cause of it is SAP AI Core doesn't return cache read and cache write information.
The actual token count usage equation should be:
Since both cache write and cache read are empty, we can set input as:
By doing this, we can get the correct context usage.
Test Procedure
Type of Change
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
Additional Notes
Important
Fixes token counting issue in
SapAiCoreHandlerfor anthropic models by recalibrating input token counts.SapAiCoreHandlerfor anthropic models by recalibrating input token counts.input = totalTokens - outputTokenswhen cache read/write are empty.sapaicore.tsto implement the recalibration logic in theSapAiCoreHandlerclass.This description was created by
for 7148f5e. You can customize this summary. It will automatically update as commits are pushed.