feat: add configurable current-request tool billing#6227
Open
bddiudiu wants to merge 1 commit into
Open
Conversation
Contributor
WalkthroughOpenAI Responses tool usage is now propagated from provider responses into built-in tool accounting, quota surcharges, configurable tool pricing, billing payloads, and per-tool usage-log displays. ChangesTool usage billing
Estimated code review effort: 4 (Complex) | ~45 minutes Sequence Diagram(s)sequenceDiagram
participant OpenAIResponses
participant OaiResponsesHandler
participant ResponsesUsageInfo
participant TextQuota
participant UsageLog
OpenAIResponses->>OaiResponsesHandler: output items and usage.tool_usage
OaiResponsesHandler->>ResponsesUsageInfo: record provisional and final tool counts
ResponsesUsageInfo->>TextQuota: built-in tool call counts
TextQuota->>UsageLog: tool_calls billing payload
UsageLog->>UsageLog: render per-tool billing rows
Possibly related PRs
Suggested reviewers: Poem
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (2)
setting/operation_setting/tools_test.go (1)
6-6: 📐 Maintainability & Code Quality | 🔵 Trivial | 💤 Low valueUse
testify/assertfor non-fatal value checks.As per coding guidelines, use
testify/assertfor non-fatal value checks and reservetestify/requirefor setup and fatal assertions.♻️ Proposed refactor
"testing" - "github.com/stretchr/testify/require" + "github.com/stretchr/testify/assert" ) func TestGetToolPriceForModelSupportsConfiguredToolNames(t *testing.T) { originalPrices := toolPriceSetting.Prices t.Cleanup(func() { toolPriceSetting.Prices = originalPrices RebuildToolPriceIndex() }) toolPriceSetting.Prices = map[string]float64{ "toutiao": 4, "toutiao:deepseek-v4-flash*": 2, "web_search:deepseek-v4-flash*": 8, } RebuildToolPriceIndex() - require.Equal(t, 2.0, GetToolPriceForModel("toutiao", "deepseek-v4-flash")) - require.Equal(t, 4.0, GetToolPriceForModel("toutiao", "other-model")) - require.Equal(t, 8.0, GetToolPriceForModel("web_search", "deepseek-v4-flash")) + assert.Equal(t, 2.0, GetToolPriceForModel("toutiao", "deepseek-v4-flash")) + assert.Equal(t, 4.0, GetToolPriceForModel("toutiao", "other-model")) + assert.Equal(t, 8.0, GetToolPriceForModel("web_search", "deepseek-v4-flash")) }Also applies to: 23-25
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@setting/operation_setting/tools_test.go` at line 6, Replace the testify/require usage in the tests with testify/assert for non-fatal value checks, including the additional occurrences at the referenced assertions; retain require only for setup failures or assertions that must stop the test immediately.Source: Coding guidelines
service/text_quota.go (1)
50-61: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick win
toolCallSurchargeItemduplicatesToolCallItemand forces a second formula computation downstream.This struct re-declares the same
name/call_count/price_per_1kJSON shape already defined byToolCallIteminservice/tool_billing.go(which additionally carriestotal_price/quota). Because the computed quota isn't stored here, the extraContent loop at Lines 377-383 has to recompute the identical price×count/1000×groupRatio×quotaPerUnit formula thataddTextToolCallSurcharge(Lines 86-107) already computed — two independent implementations of the same billing math that can silently drift.Consider either reusing
ToolCallItemdirectly, or adding an internal (non-serialized) computed-quota field set once inaddTextToolCallSurchargeand consumed by the extraContent loop.♻️ Proposed fix sketch
type toolCallSurchargeItem struct { Name string `json:"name"` CallCount int `json:"call_count"` PricePer1K float64 `json:"price_per_1k"` + quota decimal.Decimal `json:"-"` }- return decimal.NewFromFloat(pricePer1K). + quota := decimal.NewFromFloat(pricePer1K). Mul(decimal.NewFromInt(int64(callCount))). Div(decimal.NewFromInt(1000)). Mul(decimal.NewFromFloat(summary.GroupRatio)). Mul(decimal.NewFromFloat(common.QuotaPerUnit)) + summary.ToolCallItems[len(summary.ToolCallItems)-1].quota = quota + return quota- for _, toolCall := range summary.ToolCallItems { - toolQuota := decimal.NewFromFloat(toolCall.PricePer1K)... - extraContent = append(extraContent, fmt.Sprintf(..., toolQuota.String())) + for _, toolCall := range summary.ToolCallItems { + extraContent = append(extraContent, fmt.Sprintf(..., toolCall.quota.String()))🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@service/text_quota.go` around lines 50 - 61, Replace toolCallSurchargeItem with the existing ToolCallItem type from tool_billing.go, and preserve the computed quota through addTextToolCallSurcharge for the extraContent loop. Update the downstream loop to consume that stored computed value instead of reimplementing the price, count, group-ratio, and quota-per-unit calculation.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@web/default/src/features/usage-logs/components/dialogs/details-dialog.tsx`:
- Around line 337-359: Update the tool-call row construction in the details
dialog so labels are localized through t() instead of rendering toolCall.name
directly. Map backend identifiers such as web_search and file_search to their
English translation keys, and use the translated result for label while
preserving existing filtering and formatting.
---
Nitpick comments:
In `@service/text_quota.go`:
- Around line 50-61: Replace toolCallSurchargeItem with the existing
ToolCallItem type from tool_billing.go, and preserve the computed quota through
addTextToolCallSurcharge for the extraContent loop. Update the downstream loop
to consume that stored computed value instead of reimplementing the price,
count, group-ratio, and quota-per-unit calculation.
In `@setting/operation_setting/tools_test.go`:
- Line 6: Replace the testify/require usage in the tests with testify/assert for
non-fatal value checks, including the additional occurrences at the referenced
assertions; retain require only for setup failures or assertions that must stop
the test immediately.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 9e508f63-1dee-49a0-a583-1284ea333bca
📒 Files selected for processing (19)
dto/openai_response.gorelay/channel/openai/relay_responses.gorelay/channel/openai/relay_responses_test.gorelay/common/relay_info.gorelay/common/relay_info_test.goservice/text_quota.goservice/text_quota_test.gosetting/operation_setting/tools.gosetting/operation_setting/tools_test.goweb/default/src/features/system-settings/models/tool-price-settings.tsxweb/default/src/features/usage-logs/components/dialogs/details-dialog.tsxweb/default/src/features/usage-logs/types.tsweb/default/src/i18n/locales/en.jsonweb/default/src/i18n/locales/fr.jsonweb/default/src/i18n/locales/ja.jsonweb/default/src/i18n/locales/ru.jsonweb/default/src/i18n/locales/vi.jsonweb/default/src/i18n/locales/zh-TW.jsonweb/default/src/i18n/locales/zh.json
Comment on lines
+337
to
359
| if (other.tool_calls?.length) { | ||
| for (const toolCall of other.tool_calls) { | ||
| if (!toolCall.name || toolCall.call_count <= 0) continue | ||
| rows.push({ | ||
| label: toolCall.name, | ||
| value: `${toolCall.call_count}x${toolCall.price_per_1k ? ` (${fmtPrice(toolCall.price_per_1k)})` : ''}`, | ||
| }) | ||
| } | ||
| } else { | ||
| if (other.web_search && other.web_search_call_count) { | ||
| rows.push({ | ||
| label: t('Web Search'), | ||
| value: `${other.web_search_call_count}x${other.web_search_price ? ` (${fmtPrice(other.web_search_price)})` : ''}`, | ||
| }) | ||
| } | ||
|
|
||
| if (other.file_search && other.file_search_call_count) { | ||
| rows.push({ | ||
| label: t('File Search'), | ||
| value: `${other.file_search_call_count}x${other.file_search_price ? ` (${fmtPrice(other.file_search_price)})` : ''}`, | ||
| }) | ||
| if (other.file_search && other.file_search_call_count) { | ||
| rows.push({ | ||
| label: t('File Search'), | ||
| value: `${other.file_search_call_count}x${other.file_search_price ? ` (${fmtPrice(other.file_search_price)})` : ''}`, | ||
| }) | ||
| } | ||
| } |
Contributor
There was a problem hiding this comment.
📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick win
Tool-call labels bypass t(), breaking localization for this row.
label: toolCall.name renders the raw backend identifier (e.g. web_search, file_search, or any newly configured tool name) directly, unlike the else fallback which properly localizes via t('Web Search')/t('File Search'). Non-English users will see raw snake_case identifiers instead of translated labels for tool-call rows.
🌐 Proposed fix
rows.push({
- label: toolCall.name,
+ label: t(toolCall.name),
value: `${toolCall.call_count}x${toolCall.price_per_1k ? ` (${fmtPrice(toolCall.price_per_1k)})` : ''}`,
})Locale JSON files would need entries keyed by each backend tool name (e.g. "web_search", "file_search", "code_interpreter") to produce readable labels.
As per path instructions: "User-facing frontend text must use useTranslation() and t('English key')".
📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
Suggested change
| if (other.tool_calls?.length) { | |
| for (const toolCall of other.tool_calls) { | |
| if (!toolCall.name || toolCall.call_count <= 0) continue | |
| rows.push({ | |
| label: toolCall.name, | |
| value: `${toolCall.call_count}x${toolCall.price_per_1k ? ` (${fmtPrice(toolCall.price_per_1k)})` : ''}`, | |
| }) | |
| } | |
| } else { | |
| if (other.web_search && other.web_search_call_count) { | |
| rows.push({ | |
| label: t('Web Search'), | |
| value: `${other.web_search_call_count}x${other.web_search_price ? ` (${fmtPrice(other.web_search_price)})` : ''}`, | |
| }) | |
| } | |
| if (other.file_search && other.file_search_call_count) { | |
| rows.push({ | |
| label: t('File Search'), | |
| value: `${other.file_search_call_count}x${other.file_search_price ? ` (${fmtPrice(other.file_search_price)})` : ''}`, | |
| }) | |
| if (other.file_search && other.file_search_call_count) { | |
| rows.push({ | |
| label: t('File Search'), | |
| value: `${other.file_search_call_count}x${other.file_search_price ? ` (${fmtPrice(other.file_search_price)})` : ''}`, | |
| }) | |
| } | |
| } | |
| if (other.tool_calls?.length) { | |
| for (const toolCall of other.tool_calls) { | |
| if (!toolCall.name || toolCall.call_count <= 0) continue | |
| rows.push({ | |
| label: t(toolCall.name), | |
| value: `${toolCall.call_count}x${toolCall.price_per_1k ? ` (${fmtPrice(toolCall.price_per_1k)})` : ''}`, | |
| }) | |
| } | |
| } else { | |
| if (other.web_search && other.web_search_call_count) { | |
| rows.push({ | |
| label: t('Web Search'), | |
| value: `${other.web_search_call_count}x${other.web_search_price ? ` (${fmtPrice(other.web_search_price)})` : ''}`, | |
| }) | |
| } | |
| if (other.file_search && other.file_search_call_count) { | |
| rows.push({ | |
| label: t('File Search'), | |
| value: `${other.file_search_call_count}x${other.file_search_price ? ` (${fmtPrice(other.file_search_price)})` : ''}`, | |
| }) | |
| } | |
| } |
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@web/default/src/features/usage-logs/components/dialogs/details-dialog.tsx`
around lines 337 - 359, Update the tool-call row construction in the details
dialog so labels are localized through t() instead of rendering toolCall.name
directly. Map backend identifiers such as web_search and file_search to their
English translation keys, and use the translated result for label while
preserving existing filtering and formatting.
Source: Path instructions
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.
Important
📝 变更描述 / Description
针对于火山/百炼的response api请求, 实现工具调用的计费, 原平台中只能识别web_search_preview的工具调用计费.
🚀 变更类型 / Type of change
🔗 关联任务 / Related Issue
✅ 提交前检查项 / Checklist
Bug fix，我已提交或关联对应 Issue，且不会将设计取舍、预期不一致或理解偏差直接归类为 bug。📸 运行证明 / Proof of Work
(请在此粘贴截图、关键日志或测试报告，以证明变更生效)
Summary by CodeRabbit