GitLab

feat(openai): 添加 /v1/messages 端点和 API 兼容层

fix(openai): support remote compact task

fix(openai): 统一专属倍率计费链路并补齐回归测试

fix: 修复账号列表五小时用量显示为 $0.00 的问题

移除订阅扣费 stub 中未被使用的状态字段与赋值，
消除 golangci-lint 的 unused 告警，保持回归测试语义不变。
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

抽取共享的用户分组专属倍率解析器，统一缓存、singleflight 与回退逻辑。\n\n让 OpenAI 独立计费链路复用专属倍率解析，修复 usage 记录与实际扣费未命中用户专属倍率的问题。\n\n补齐 OpenAI 计费与解析器单元测试，并修复全量回归中暴露的 lint 阻塞项。\n\nCo-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

commit 80ae592c 引入 lite 模式优化首次加载性能，但将窗口费用查询也一起跳过了。
commit 491a7444

 尝试用 30 秒快照缓存修复，但缓存过期后问题复现。

移除窗口费用查询的 lite/非 lite 区分，始终执行 PostgreSQL 聚合查询。
同时删除不再需要的 account_window_cost_cache.go 文件。
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

在 admin.proxies 下添加 protocols 对象的中文翻译，
与 en.ts 保持一致。
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

- Add recoverAnthropicMessagesPanic for Messages handler to return
  Anthropic-formatted errors instead of OpenAI Responses format on panic
- Add nil check for rateLimitService.HandleUpstreamError in
  ForwardAsAnthropic to match defensive pattern used elsewhere
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Add Anthropic Messages API support for OpenAI platform groups, enabling
clients using Claude-style /v1/messages format to access OpenAI accounts
through automatic protocol conversion.

- Add apicompat package with type definitions and bidirectional converters
  (Anthropic ↔ Chat, Chat ↔ Responses, Anthropic ↔

 Responses)
- Implement /v1/messages endpoint for OpenAI gateway with streaming support
- Add model mapping UI for OpenAI OAuth accounts (whitelist + mapping modes)
- Support prompt caching fields and codex OAuth transforms
- Fix tool call ID conversion for Responses API (fc_ prefix)
- Ensure function_call_output has non-empty output field
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

fix(openai): passthrough accounts bypass model mapping check

透传模式账号仅替换认证，应允许所有模型通过。之前调度阶段的
isModelSupportedByAccount 不感知透传模式，导致 model_mapping
中未配置的新模型（如 gpt-5.4）被拒绝返回 503。

feat(openai): 增加 GPT-5.4 支持并修复长上下文计费与白名单回归

lite 模式下从快照缓存读取窗口费用，非 lite 模式查询后写入缓存

补齐 gpt-5.4 fallback 的长上下文计费元信息，\n确保超过 272000 输入 token 时对整次会话应用\n2x 输入与 1.5x 输出计费规则。\n\n同时将官方快照 gpt-5.4-2026-03-05 加入前端\n白名单候选与回归测试，避免 whitelist 模式误拦截。\n\nCo-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

(cherry picked from commit d95497af87f608c6dadcbe7d6e851de9413ae147)

- 接入 gpt-5.4 模型识别与规范化，补充默认模型列表
- 增加 gpt-5.4 输入/缓存命中/输出价格与计费兜底逻辑
- 同步前端模型白名单与 OpenCode 上下文窗口（1050000/128000）
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
(cherry picked from commit 924476dcac6181cd0f3ee731ec7b73672ff03793)

feat: add account load_factor for scheduling load calculation

feat: add configurable spending limit for API Key accounts

LoadFactor should only affect scheduling weight, not load rate reporting.

Replace `<` comparisons with Math.max/ternary+>= to avoid Vue template
parser treating `<` as HTML tag start in attribute values.

- Fix bulk edit: send 0 instead of null/NaN to clear load_factor
- Fix edit modal: explicit NaN check instead of implicit falsy
- Fix create modal: use ?? instead of || for load_factor
- Add load_factor upper limit validation (max 10000)
- Add //go:build unit tag and self-contained intPtrHelper in test
- Add design intent comments on WaitPlan.MaxConcurrency

- Normalize non-standard status (e.g. "error") to "active" on edit load
- Add pre-submit validation for status field to prevent 400 errors
- Update load factor hint: "提高负载因子可以提高对账号的调度频率"

The test could flake depending on exact execution time near midnight
boundaries. Use a range check (29 or 30) instead of exact equality.

- Extract postUsageBilling() to consolidate billing logic across
  GatewayService.RecordUsage, RecordUsageWithLongContext, and
  OpenAIGatewayService.RecordUsage, eliminating ~120 lines of
  duplicated code
- Fix account quota to use TotalCost × accountRateMultiplier
  (was using raw TotalCost, inconsistent with account cost stats)
- Fix RecordUsageWithLongContext API Key quota only updating in
  balance mode (now updates regardless of billing type)
- Fix WebSocket client disconnect detection on Windows by adding
  "an established connection was aborted" to known disconnect errors

Antigravity APIKey accounts were incorrectly routed to
testAntigravityAccountConnection which calls AntigravityTokenProvider,
but the token provider only handles OAuth and Upstream types, causing
"not an antigravity oauth account" error.

Extract routeAntigravityTest to route APIKey accounts to native
Claude/Gemini test paths based on model prefix, matching the
gateway_handler routing logic for normal requests.

- Extract quota limit card/toggle UI into QuotaLimitCard.vue component
- Use v-model pattern for clean parent-child data flow
- Integrate into both EditAccountModal and CreateAccountModal
- All apikey accounts (all platforms) now support quota limit on creation
- Bump version to 0.1.90.6

- Redesign quota limit section with card layout and toggle switch
- Add watch to clear quota value when toggle is disabled
- Add i18n keys for toggle labels and hints (zh/en)
- Bump version to 0.1.90.5

- Add configurable spending limit (quota_limit) for apikey-type accounts
- Atomic quota accumulation via PostgreSQL JSONB operations on TotalCost
- Scheduler filters out over-quota accounts with outbox-triggered snapshot refresh
- Display quota usage ($used / $limit) in account capacity column
- Add "Reset Quota" action in account menu to reset usage to zero
- Editing account settings preserves quota_used (no accidental reset)
- Covers all 3 billing paths: Anthropic, Gemini, OpenAI RecordUsage

chore: bump version to 0.1.90.4