feat(stats): add model cost estimation to /stats model (#3780)#219
Merged
Conversation
Cherry-picks upstream qwen-code PR QwenLM#3780. Skips the source-attribution features (`bySource`, `flattenModelsBySource`, MAIN_SOURCE) that came along for the ride — those depend on a separate un-ported subagent attribution PR. What's in: - `costCalculator` util — pure function, takes input/output tokens + per-model `ModelPricing` ({inputPerMillionTokens, outputPerMillionTokens}), returns null when pricing is absent or total cost is zero. - `ui.modelPricing` setting — a `Record<string, ModelPricing>` keyed by model name. When set, the `/stats model` TUI shows a Cost section, and `/stats model` in non-interactive mode appends "Estimated cost" per model. - `ModelStatsDisplay` adds a Cost section after Tokens, gated on at least one model having computable cost (`hasPricing`). - `statsCommand` model subcommand: in non-interactive/ACP modes, returns a text message including per-model Estimated cost lines. Conflict-resolution notes: - Dropped `supportedModes` field — un-ported metadata, single consumer, no enforcement in our slash system. - Dropped upstream's expanded non-interactive output for top-level /stats and /stats tools (depends on `metrics.files.totalLinesAdded` which our SessionMetrics doesn't have). Kept just the model-cost path, which is the headline feature. - Rewrote `ModelStatsDisplay` cost row from scratch against our HEAD shape (no `bySource`, no `key.split('::')`) instead of taking upstream's full re-render. - Gate non-interactive path on explicit `=== 'non_interactive' || === 'acp'` rather than `!== 'interactive'`, so undefined defaults to interactive and existing tests keep working. - vscode-ide-companion settings.schema.json: deleted in our fork, removed from cherry-pick. Tests: 32 new tests pass (costCalculator, statsCommand model cost in non-interactive, ModelStatsDisplay cost row). Broader sweep of 1186 tests across packages/cli/src/ui/{commands,components} all pass. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
|
Warning Rate limit exceeded
To keep reviews running without waiting, you can enable usage-based add-on for your organization. This allows additional reviews beyond the hourly cap. Account admins can enable it under billing. ⌛ How to resolve this issue?After the wait time has elapsed, a review can be triggered using the We recommend that you space out your commits to avoid hitting the rate limit. 🚦 How do rate limits work?CodeRabbit enforces hourly rate limits for each developer per organization. Our paid plans have higher rate limits than the trial, open-source and free plans. In all cases, we re-allow further reviews after a brief timeout. Please see our FAQ for further information. ℹ️ Review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (7)
✨ Finishing Touches🧪 Generate unit tests (beta)
Review rate limit: 0/5 reviews remaining, refill in 5 minutes and 44 seconds. Comment |
Code Coverage Summary
CLI Package - Full Text ReportCore Package - Full Text ReportFor detailed HTML reports, please see the 'coverage-reports-22.x-ubuntu-latest' artifact from the main CI run. |
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
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.
Summary
/stats model(TUI cost row + non-interactive Estimated cost lines)bySource/flattenModelsBySource/MAIN_SOURCE) — those depend on a separate un-ported subagent attribution PRmetrics.files.totalLinesAddedwhich our SessionMetrics doesn't haveWhat's in
costCalculatorutil — pure function, takes input/output tokens +ModelPricing({inputPerMillionTokens, outputPerMillionTokens}); returns null when pricing absent or total cost is zeroui.modelPricingsetting —Record<string, ModelPricing>keyed by model name. When set, the TUI shows a Cost section and the non-interactive/stats modelappends Estimated cost per modelModelStatsDisplayadds a Cost section after Tokens, gated onhasPricingstatsCommandmodel subcommand non-interactive/ACP path returns text with per-model Estimated cost linesConflict-resolution notes
supportedModesfield — un-ported metadata, no enforcement in our slash systemModelStatsDisplaycost row from scratch against our HEAD shape (nobySource, nokey.split('::')) instead of taking upstream's full re-render=== 'non_interactive' || === 'acp'rather than!== 'interactive', so undefined defaults to interactive (existing tests still pass)Configuration
Add to settings.json:
```json
{
"ui": {
"modelPricing": {
"qwen3-coder-plus": {
"inputPerMillionTokens": 0.30,
"outputPerMillionTokens": 1.20
}
}
}
}
```
Test plan
🤖 Generated with Claude Code