fix: calculate OpenRouter costs locally to avoid API miscalculation #8651
+80
−2
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
![roomote[bot]](https://avatars.githubusercontent.com/in/1546624?s=80&v=4){kind=small}
- Calculate costs using model pricing info instead of relying on OpenRouter API response - Fixes issue where OpenRouter returns incorrect cost values (e.g., $0.46 instead of $1.50+ for 527k tokens) - Falls back to API cost if model pricing info is unavailable - Added test case to verify fix for reported issue #8650
dosubot
bot
added
size:M
![ellipsis-dev[bot]](https://avatars.githubusercontent.com/in/64358?s=60&v=4){kind=small}
| // Use locally calculated cost, but fall back to API response if our calculation fails | ||
| // or if model pricing info is not available | ||
| const apiCost = (lastUsage.cost_details?.upstream_inference_cost || 0) + (lastUsage.cost || 0) | ||
| const totalCost = modelInfo.inputPrice && modelInfo.outputPrice ? localCost : apiCost |
There was a problem hiding this comment.
Using modelInfo.inputPrice && modelInfo.outputPrice to decide whether to use locally calculated cost may misinterpret valid zero prices as missing. Consider checking explicitly for undefined/null (e.g. modelInfo.inputPrice != null && modelInfo.outputPrice != null) to ensure that a price of 0 isn’t treated as false.
Suggested change
| const totalCost = modelInfo.inputPrice && modelInfo.outputPrice ? localCost : apiCost | |
| const totalCost = modelInfo.inputPrice != null && modelInfo.outputPrice != null ? localCost : apiCost |
![roomote[bot]](https://avatars.githubusercontent.com/in/1546624?s=60&v=4){kind=small}
| // Use locally calculated cost, but fall back to API response if our calculation fails | ||
| // or if model pricing info is not available | ||
| const apiCost = (lastUsage.cost_details?.upstream_inference_cost || 0) + (lastUsage.cost || 0) | ||
| const totalCost = modelInfo.inputPrice && modelInfo.outputPrice ? localCost : apiCost |
There was a problem hiding this comment.
The truthy check here will incorrectly fall back to API cost when model prices are 0 (e.g., free tier models). If inputPrice or outputPrice is 0, the condition evaluates to false even though 0 is a valid price.
Suggested change
| const totalCost = modelInfo.inputPrice && modelInfo.outputPrice ? localCost : apiCost | |
| const totalCost = modelInfo.inputPrice !== undefined && modelInfo.outputPrice !== undefined ? localCost : apiCost |
hannesrudolph
added
the
Issue/PR - Triage
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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 GitHub Issue
Closes: #8650
Roo Code Task Context (Optional)
This PR was created with assistance from Roo Code.
Description
This PR fixes the cost miscalculation issue when using Claude 3.5 Sonnet (and other models) through OpenRouter. The issue occurred because OpenRouter's API was returning incorrect cost values in the usage response.
Key implementation details:
calculateApiCostOpenAIutility function for consistent cost calculation across the codebaseWhat reviewers should pay attention to:
src/api/providers/openrouter.tslines 199-226Test Procedure
How I tested:
How reviewers can verify:
cd src && npx vitest run api/providers/__tests__/openrouter.spec.tsPre-Submission Checklist
Screenshots / Videos
Not applicable - this is a backend cost calculation fix with no UI changes.
Documentation Updates
The fix is transparent to users and doesn't change any APIs or user-facing behavior.
Additional Notes
This fix ensures that users see accurate cost calculations when using OpenRouter, particularly for high token count sessions. The fallback mechanism ensures backward compatibility if model pricing information is not available.
Get in Touch
Automated PR created via GitHub API
Important
Fix cost miscalculation in
OpenRouterHandlerby calculating costs locally using model pricing info and adding fallback to API costs.OpenRouterHandlerinopenrouter.tsto calculate costs locally using model pricing info.calculateApiCostOpenAIfor consistent cost calculation.openrouter.spec.tsto validate local cost calculation for 527k input tokens.createMessage()inopenrouter.tslines 199-226.This description was created by
for 1eecf1f. You can customize this summary. It will automatically update as commits are pushed.