fix(cost): never fabricate spend — flat-rate engines $0, cache-aware …#204
Merged
Conversation
…pricing, budget checks on metered cost only Root cause of the '$8.79 actual vs $0.07 estimated' budget warning on a kimi coding-plan run: (1) kimi-for-coding/minimax-/zai-coding-plan had no price-table entry and fell to a made-up $2/Mtok default although their marginal cost is $0 (flat-rate subscription); (2) AI-SDK promptTokens include cache reads, so a multi-turn tool loop re-priced the whole context every call (~4.4M 'tokens'); (3) the plan budget check and plan-mode planning watchdog read the ballpark totalCostUsd that token-tracker itself documents as 'NOT a bill'. - isFlatRateEngine (naming-convention heuristic until EngineDefinition carries billing): estimateCost returns 0; sdk usage on flat-rate engines counts as unmetered in getStats. - estimateCostCacheAware: cached input priced at the 10% cache-read rate, clamped against promptTokens only (never discounts output tokens). - plan-mode snapshotTokens + session planning-budget warning read meteredCostUsd. - Review round (claude/codex/agy): both codex findings fixed (planning watchdog ballpark, clamp vs promptTokens). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
…pricing, budget checks on metered cost only
Root cause of the '$8.79 actual vs $0.07 estimated' budget warning on a kimi coding-plan run: (1) kimi-for-coding/minimax-/zai-coding-plan had no price-table entry and fell to a made-up $2/Mtok default although their marginal cost is $0 (flat-rate subscription); (2) AI-SDK promptTokens include cache reads, so a multi-turn tool loop re-priced the whole context every call (~4.4M 'tokens'); (3) the plan budget check and plan-mode planning watchdog read the ballpark totalCostUsd that token-tracker itself documents as 'NOT a bill'.