Commit Graph
1744 Commits
Author SHA1 Message Date
feeeei 8b74741efe fix(groups): 广场 Vary 保留 Origin,分组译文语言代码统一大小写
- 模型广场以追加方式写入 Vary: Accept-Language,保留 CORS 中间件为具体来源
  写入的 Vary: Origin
- 管理端写入分组译文时语言代码统一为 BCP 47 惯用大小写(en、zh-TW),
  统一后重复的语言代码(如 en 与 EN)返回 400 INVALID_GROUP_I18N
- 用户侧按语言取译文时规范写法的键优先,仅大小写不同的多个键按字节序取最小者,
  同一请求语言的结果固定
2026-10-12 18:09:03 +08:00
feeeei 9a51333101 feat(groups): 分组名称与描述支持多语言
- groups 新增 i18n JSONB 列(迁移 242):语言代码 -> {name, description},只存译文;
  name/description 原列语义不变,仍用于唯一性校验、搜索与各类日志
- 用户侧接口由后端按请求语言输出:取 Accept-Language 中权重最高的一项,先按完整
  语言代码、再按主语言匹配;该语言有译文的字段换成译文,否则保持原字段,且不再带出
  译文表。覆盖密钥、可用分组、订阅(含进度与汇总)、兑换、使用记录、可用渠道、
  模型广场、套餐列表
- 管理端接口一律返回原字段;分组管理接口额外带 i18n 供编辑
- 创建/编辑分组:名称、描述的标题行右侧新增语言下拉,同一个输入框在默认内容与各语言
  版本之间切换;语种跟随前端 availableLocales
- 管理端写入:创建/更新接受 i18n(更新时缺省不修改,传入则整体替换),校验语言代码
  形态与名称长度并丢弃空译文;简易模式同样放行;复制分组只带走描述译文
- 切换界面语言后整页刷新,让已加载的数据换成新语言;模型广场响应加
  Vary: Accept-Language
- 网关路径,以及只带 group_name 字符串的统计、错误记录出口不做本地化
2026-10-09 11:39:18 +08:00
SavitarC 9264414612 fix(provider): 修复 Claude 计费与钱包冷却并隔离模型目录缓存 2026-10-09 10:11:06 +08:00
SavitarCandClaude Opus 5.5 c556aee8fd refactor(platform): 多协议供应商的转发与探测判断改查 profile
平台清单之外,后端还有几处用"国产厂商"或"国产厂商 ∪ OpenCode"表示
多协议供应商,新登记的供应商会被漏掉(例如上游模型同步直接报不支持)。

- 适用于全部多协议供应商的判断改用 IsMultiProtocolAPIKey:上游模型同步与
  模型目录基址、Anthropic count_tokens 本地估算、HTML 403 豁免与累计计数、Claude 型号
  计费候选过滤;
- 新增 RoutesProtocolByInbound(profile 分流方式为按入站协议),替换原来
  只针对国产厂商、实际描述"按入站协议分流"的判断:Responses 直转回退、
  Responses 能力标记(含创建 / 更新后的异步探测调度)、上游计费探测的
  adaptive 基址、Anthropic API Key 鉴权方式、OpenAI 输入 token 本地估算。
  按模型分流的供应商(OpenCode)仍走 routesByModel 分支。

依赖各家额度 / 余额接口的逻辑(Coding Plan 窗口、余额不足冷却等)仍按平台
显式判断。现有平台上两组判断与改动前逐一相同(测试覆盖全部平台与非法值);
另测试临时登记的新供应商按分流方式分别获得国产厂商与 OpenCode 的行为。

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-09 10:10:33 +08:00
SavitarCandClaude Opus 5.5 3b89cc08dd refactor(platform): 后端平台列表改由平台清单派生
新增 domain/platforms.go 作为具体平台的单一数据来源:每个平台一条登记,
写明展示名、所走网关(anthropic / openai / gemini / antigravity)、是否国产
单一厂商、定价目录 provider 名,顺序即展示顺序。

以下列表与判断改为查平台清单:
- "全部具体平台":用户平台额度白名单、错误透传规则平台列表、组合路由目标
  校验、上游计费探测资格、调度快照遍历、组合分组的定价 / 模型映射回退、
  组合分组默认模型列表;
- "OpenAI 网关族":IsOpenAICompatible、NormalizeOpenAICompatiblePlatform、
  Responses 兼容路由判断、count_tokens 分发、组合分组的 OpenAI 文本目标;
- 国产厂商判断、渠道定价同步的 provider 映射;
- 多协议 API Key 供应商相关的请求头覆盖资格、凭证测试、OpenAI 基址回退改用
  IsMultiProtocolAPIKeyProvider;
- 分组平台与组合路由目标的请求校验改为自定义校验标签 group_platform /
  concrete_platform,拒绝规则不变(大小写、首尾空格、别名一律拒绝)。

组合分组的逐平台回退是先命中者生效,CompositePrecedencePlatformIDs 保持
原有顺序 anthropic、gemini、openai,其余按登记顺序追加。依赖具体实现的
能力(OAuth、渠道监控、调度阈值、余额 / 用量窗口、Codex 模型目录等)仍在
各自代码中显式列出。

等价性测试把改动前的每个列表与 switch 原样保留为基准,逐一比较派生结果
(含顺序),谓词覆盖全部平台、composite 与非法值 × 全部账号类型。

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-09 10:10:33 +08:00
Wesley Liddick 8e2c4fad32 Merge pull request #7793 from wucm667/fix/issue-7784-composite-route-model-list
fix(models): list composite exact route aliases
2026-10-08 11:39:51 +08:00
Wesley Liddick bc63bd2874 Merge pull request #7782 from lyy0709/codex/fix-remote-model-catalog
fix(models): enforce account mappings for remote Codex discovery
2026-10-08 11:33:27 +08:00
Wesley Liddick 3a7fd4ae78 Merge pull request #7837 from yJader/fix/antigravity-account-test-models
fix: use account mapping keys for Antigravity connectivity tests
2026-10-08 11:32:59 +08:00
Wesley Liddick 6db4171cfb Merge pull request #7779 from SPHINX998/fix/ws-turn-group-pricing
fix(openai-ws): bill each later turn with the current group pricing
2026-10-08 11:25:10 +08:00
yJader 856d7ea905 fix: expose mapped antigravity test models 2026-10-03 20:56:49 +08:00
wucm667 5800abf1cb fix(models): preserve composite catalog isolation after base update 2026-10-03 09:11:53 +08:00
Wesley Liddick 2529371bab Merge pull request #7676 from william-drakemond/fix/email-code-token-atomic
fix(email): 验证码尝试次数原子化 + 重置密码 token 哈希存储且原子单次消费
2026-10-02 14:32:42 +08:00
Wesley Liddick 5637f469fa Merge pull request #7802 from feeeei/feat/recharge-promotion-tiers
feat(payment): 充值优惠阶梯,支持赠金与折扣两种模式
2026-10-02 14:30:29 +08:00
shaw 9eb394e3e9 fix(typesafe): 堵住校验解析差异与审计盲区,收紧模型列表口径
- System One 校验拒绝重复键与大小写/Unicode 折叠变体键(请求顶层、
  questions、问题对象):encoding/json 大小写不敏感且取最后一个重复键,
  而原始 body 原样转发,此前 {"model":"x","MODEL":"jev-latest"} 可绕过
  jev-latest 限制、模型白名单与流式限制。
- Prompt 审计与旧审核同时收集对象键名、问题 ID 与未知扩展字段(不含规范
  字段名与已校验的 type),此前把内容放进键名即可完全绕过审计。
- 上游成功响应的 usage/model 宽松解码(浮点、数字字符串),避免上游已
  计费而网关整体 502 不记账;解码失败记录 response_error ops 事件。
- 上游 413 与 400/422 同视为请求错误:透传状态码、不切换、不改账号状态。
- TypeSafe base_url 末尾的 /v1 自动剥离,避免拼出 /v1/v1/systemone;
  /v1/systemone 改挂文本请求体上限。
- Composite 静态兜底模型列表、管理端候选列表恢复为不含 jev-latest;
  Codex 清单永不列出 TypeSafe 模型;TypeSafe 分组候选默认 jev-latest。
- 创建账号从 Grok 切到 TypeSafe 时重置为白名单模式;错误透传平台列表补 typesafe。
2026-10-02 13:51:58 +08:00
shaw 2d8a386811 fix(typesafe): 补齐平台迁移、审计覆盖、端点隔离与错误策略
- 新增 241 迁移:user_platform_quotas / composite_model_routes 的 CHECK 约束
  加入 typesafe。此前设置 typesafe 默认配额会让注册时的多行配额快照整体
  违约(fail-open 后新用户所有平台配额丢失),单用户配额与 Composite 路由保存 500。
- Prompt 审计新增 typesafe_systemone 协议提取(state + 各问题 instructions/
  criteria/选项标签,键排序保证哈希稳定);此前提取为空,阻断与异步审计均被放行。
  旧内容审核同样覆盖问题文本,且不再丢弃含 <system-reminder> 的 System One 文本。
- TypeSafe 分组及路由到 TypeSafe 的 Composite 请求访问 Messages / count_tokens /
  Chat Completions / Responses 时返回 404,避免以 x-api-key 打到错误上游路径并
  污染账号状态;TypeSafe 账号 base_url 为空时不再回退 api.anthropic.com。
- SystemOne 补在途余额预留、利润控制准入终检、wrapReleaseOnDone,失败切换改用
  共享 FailoverState(同账号重试 / 池模式 / 临时封禁)。
- 上游错误沿用共享账号错误策略:402/403、自定义错误码、临时不可调度规则生效,
  记录 ops 上游错误;400/422 仍不切换且永不改变账号状态。
- 账号测试改走原生 System One 探测,可用模型只返回 jev-latest;
  响应超限显式报错,响应 Content-Type 仅透传 JSON 类型。
2026-10-02 13:08:40 +08:00
aofee 7225538358 merge: sync Jev support with latest upstream main 2026-10-02 09:37:30 +08:00
feeeei 1aa34d478d feat(payment): 充值优惠阶梯支持折扣模式,快捷金额角标改为促销价签
- 新增全局模式开关 RECHARGE_BONUS_MODE(bonus / discount),接入支付配置、后台设置接口与 checkout-info
- 折扣模式:到账不变,实付基数按百分比打折(按币种精度取整),手续费 / 每日限额 / 渠道选实例沿用折后基数;
  百分比须 < 100,运行时 ≥ 100 按无优惠处理
- 充值页角标改为单行红色价签(赠金「+20%」/ 折扣「20% OFF」),第二行按模式显示到账或折后实付;
  摘要卡折扣模式显示优惠行,渠道限额校验改用折后实付
- 后台「充值优惠阶梯」拆为独立卡片:赠金 / 折扣切换、紧凑档位表格、区间预览、活动文案
- 补前后端单测与 zh/en 文案
2026-10-02 00:25:41 +08:00
feeeei 1b1039f4fd feat(payment): 充值赠送阶梯(按金额档位额外赠送到账额度)
- 后台支付设置新增「充值赠送阶梯」(满 X 送 Y%,RECHARGE_BONUS_TIERS)与 Markdown 活动文案
  (RECHARGE_BONUS_NOTICE):阈值按用户输入的支付金额命中(取不超过金额的最大档),
  赠送按到账基数(输入 × 充值倍率)计算;严格校验写入、宽松解析读取,订阅订单不参与
- 下单时按当时配置计算赠送并落库:payment_orders 新增 bonus_amount 列(迁移 241),
  amount 仍为到账总额,到账/兑换码/退款逻辑不变;推广返利基数改为 amount - bonus_amount
- checkout-info 下发 recharge_bonus_tiers / recharge_bonus_notice;创建订单与订单查询响应带 bonus_amount
- 充值页:快捷金额按钮右上角「+N%」角标 + 第二行到账金额,金额卡顶部渲染活动文案,
  订单摘要新增赠送额度行并在有赠送时始终显示到账余额;订单列表/详情/支付结果/成功面板显示含赠送
- 后台新增 RechargeBonusTierEditor(行内校验重复/非法阈值,自动排序并显示区间预览)
- 补后端单测、契约测试与前端 vitest,zh/en 文案
2026-10-02 00:25:41 +08:00
wucm667 1888de135d fix(models): list composite exact route aliases 2026-10-01 22:54:05 +08:00
lyy0709 71e97d1353 fix(models): enforce account mappings for remote Codex discovery 2026-10-01 03:30:56 -07:00
SPHINX998andClaude Opus 5.5 73381fc9c7 fix(openai-ws): bill each later turn with the current group pricing
The Responses WebSocket handler reads the API key auth snapshot once at
connect and reuses it for every turn: RecordUsage takes the group pricing
from that snapshot, and the profit gate reinstalled in BeforeTurn takes
its selling price from the connect-time ctxkey.Group. A group price
change therefore never reaches an open connection until the client
reconnects, while HTTP requests pick it up once the auth cache entry is
invalidated.

Each later turn now re-reads the key through APIKeyService.GetByKey (the
auth cache the middleware uses) in BeforeTurn. When the key ID, group ID,
group platform and subscription type are unchanged, the fresh group is
swapped into ctxkey.Group for the turn's profit gate and passed to that
turn's RecordUsage. A failed lookup or a key moved to another group keeps
the connection snapshot.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-01 11:27:35 +08:00
Wesley Liddick 96f4c115c9 Merge pull request #7726 from william-drakemond/feat/claude-reset-redeem
feat(claude): 手动兑换 Claude 重置次数 / Manually redeem Claude limit resets
2026-09-30 14:33:23 +08:00
Wesley Liddick 327c322184 Merge pull request #7730 from lyy0709/codex/gpt61-sol-codex-plans
feat(openai): support GPT-6.1 Sol, Codex plans and Astra Ultrafast
2026-09-30 14:32:59 +08:00
Wesley Liddick 4f0c079e27 Merge pull request #7681 from william-drakemond/fix/balance-inflight-reservation
fix(billing): Redis 在途预留,防止余额模式并发透支
2026-09-30 14:31:04 +08:00
Wesley Liddick a0f41f95a0 Merge pull request #7579 from feeeei/fix/claude-code-only-fallback-openai-compat
fix(gateway): Claude Code 限制分组的 OpenAI 兼容入口改走降级分组,不再直接 403
2026-09-30 10:47:08 +08:00
lyy0709 9688571a83 feat(models): support GPT-6.1 Sol 2026-09-29 11:49:26 -07:00
196b47757a feat(claude): redeem native limit resets with server-selected grant
POST /admin/accounts/:id/claude/reset-credits/redeem (Idempotency-Key
required, same middleware chain as the Codex reset-quota route). The server
re-queries eligibility and claims only the upstream next_grant_id when it is
redeemable; grant/org IDs never reach the client. Claims are serialized by
account and organization Redis leases, and a durable per-organization fence
in the idempotency table blocks further redemption after an unconfirmed
outcome (settles after 24h). Ported from upstream PR #7591.

Co-authored-by: korkin25 <eugeny.kuzakov@gmail.com>
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-09-29 11:41:33 -04:00
Wesley Liddick fe9dadfc7d Merge pull request #7679 from PMExtra/feat/risk-control-allowlist
feat: 新增可信用户风控白名单,保留审计并免除本地自动处罚
2026-09-29 15:53:18 +08:00
Wesley Liddick 052e08e764 Merge pull request #7684 from william-drakemond/feat/claude-reset-credits-ui
feat(claude): 查询 Claude 重置次数(只读),UI 复用 Codex 次数按钮 / Read-only Claude reset credit count, reusing Codex count UI
2026-09-29 15:50:53 +08:00
Wesley Liddick ccb027b515 Merge pull request #7378 from wucm667/fix/issue-7368-composite-ws-routing
fix(gateway): resolve composite routes for websocket aliases
2026-09-29 15:43:52 +08:00
korkin25 6a8f921650 feat(claude): expose sanitized native reset credit status 2026-09-28 18:04:13 -04:00
William DrakeandClaude Opus 5.5 52c123e6f4 fix(billing): hold inflight reservation until billing lands; price estimate like billing; wire remaining endpoints
Address gaps found in adversarial verification of the in-flight balance
reservation:

- Early release: the reservation is now a ref-counted handle carried on the
  request context. Billing tasks submitted via submit*UsageRecordTask take a
  ref and return it only after RecordUsage has run; with reservation enabled
  the balance cache is deducted synchronously inside the billing task, so the
  reservation is released only after admission can see the deduction. Dropped
  tasks return their ref immediately. The reservation is renewed every ttl/3
  while the handler runs; after the handler returns it is no longer renewed,
  so the post-handler hold is bounded by one TTL.
- Estimate: use the same billing model (channel mapping + BillingModelSource,
  max over requested/mapped when ambiguous) and ModelPricingResolver (group
  ModelPricing incl. wildcards, channel per_request/image/video/token
  overrides), default/group/user rate + peak multiplier, plus image, video,
  audio and search surcharge kinds. Unpriced requests fail open with a
  throttled log; new billing.inflight_reservation.fail_closed_on_unpriced
  (default false) rejects them instead.
- Wire images, embeddings, web/x search, alpha search, grok media generation,
  grok realtime/voice, OpenAI-platform /v1/messages and ResponsesWebSocket
  (session-level reservation with renewal).

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-09-28 16:14:43 -04:00
William DrakeandClaude Opus 5.5 43f88c99aa fix(billing): reserve in-flight balance to prevent concurrent overdraft
Balance-mode admission only checked cached balance > threshold, so N
concurrent expensive requests could all pass and overdraw the user.
Add a Redis in-flight reservation: at admission estimate a conservative
per-request cost (input estimate + output price x max_tokens, times rate
multiplier) and atomically require cached_balance - inflight >= estimate
(first in-flight request always admitted). Reservations are released via
defer on every path and expire by TTL. Fails open on estimation / Redis
errors. Configurable via billing.inflight_reservation (enabled by default).

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-09-28 14:20:49 -04:00
PMExtra 6a69dd051a feat: add risk control user allowlist 2026-09-29 01:50:21 +08:00
William DrakeandClaude Opus 5.5 b1ce3c6d59 fix(email): atomic verify-code attempts and hashed single-use reset tokens
- Verification code attempts (register + notify email) are now reserved via
  an atomic Redis INCR (Lua) before comparing, so concurrent wrong guesses
  cannot exceed the 5-attempt cap.
- Password reset tokens are stored as SHA-256 hashes and consumed with an
  atomic Lua compare-and-delete, so a token can only be used once even under
  concurrent requests.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-09-28 12:26:28 -04:00
wucm667 aa3a3ea787 feat(dashboard): toggle recent usage between tokens and spending 2026-09-28 14:48:42 +08:00
Wesley Liddick b3c7038cf4 Merge pull request #7526 from HuntercodeT/fix/passthrough-model-discovery-merge
fix(gateway): let OpenAI passthrough accounts supplement model discovery instead of hiding mapped models
2026-09-28 10:49:01 +08:00
Wesley Liddick 79c18ec836 Merge pull request #7609 from feeeei/fix/client-disconnect-499
fix(gateway): 客户端断开统一标记 499,不再被记成 502/200 上游错误
2026-09-28 10:04:07 +08:00
feeeei f4f8ff04db fix(gateway): 客户端断开统一标记 499,不再被记成 502/200 上游错误
客户端在响应提交前断开时,只有排队/换号阶段和 OpenAI 的兜底写入会标 499。
Claude 非流式与 Gemini 的 Chat Completions / Messages / Responses 兼容入口
在等待上游时断开,由 GatewayHandler.ensureForwardErrorResponse 补写 502;
Gemini 原生入口什么都不写,落成 200 并被 ops 记成一条 provider 的
"Recovered upstream error";Antigravity 回写 502。这些行都按上游失败计入
SLA 与上游错误数。

- GatewayHandler.ensureForwardErrorResponse 补客户端断开检查,与 OpenAI 版一致
- GatewayHandler 的 Messages / ChatCompletions / Responses 与 Gemini 原生入口:
  首次选号失败时先判断客户端断开;ChatCompletions / Responses 循环开头改用
  failoverClientGone,不再直接 return 落成 200
- Gemini 原生入口:forward 失败时补 499 标记;并发槽错误改走
  concurrencyErrorResponse 统一映射(断开 499、排队满/超时 429、其余 503)
- Antigravity 上游响应前断开回写 499;googleapi 与 Antigravity 的 Google 错误
  格式把 499 映射为 CANCELLED
- OpenAI alpha search 非 failover 错误分支补断开检查
- Claude / Gemini / OpenAI 传输层错误:请求 context 已取消时不记 ops 上游事件
- ops:IgnoreContextCanceled 开启时跳过没有上游错误上下文的 499;带上游错误的
  499(上游失败后客户端没等到换号结果就离开)仍按上游失败落库

计费不变:Gemini 流式中途断开的用量处理不在本次范围。新增 handler/service
单测覆盖各入口的断开场景与 ops 过滤规则。
2026-09-26 17:13:10 +08:00
Achord Chan 510b680fca Merge upstream main and retain upstream scheduler test fix 2026-09-26 15:47:52 +08:00
feeeei b31b435091 fix(gateway): Claude Code 限制分组的 OpenAI 兼容入口改走降级分组,不再直接 403
分组开启 claude_code_only 并配置降级分组时,/v1/messages 会在选号入口的
checkClaudeCodeRestriction 中把非 Claude Code 请求解析到降级分组;但
GatewayHandler.ChatCompletions / Responses 在 handler 层只要分组开了
claude_code_only 就直接返回 403,不看降级分组,请求到不了调度层:只会发
Chat Completions / Responses 的第三方客户端拿这类分组的 Key 一律 403。

- handler 层提前拒绝只在未配置降级分组时生效,拒绝的状态码与文案不变
- 配置了降级分组时请求继续,由 SelectAccountWithLoadAwareness 解析到降级分组
  选号,再经已有的 Chat Completions / Responses ⇄ Anthropic Messages 转换转发;
  service 层无改动
- 与 /v1/messages 的降级语义一致:选号与渠道限制预检按降级分组,计费与用量
  记录仍按 API Key 所属分组

新增 handler 单测覆盖两个入口在有/无降级分组时的行为。
2026-09-24 16:20:28 +08:00
wucm667 13fcfc0f3b fix(gateway): resolve composite routes for websocket aliases 2026-09-23 15:29:11 +08:00
alfadb d0731be2f0 fix(opencode): lite 列表 DTO 补 opencode_go_usage,修复列表页显示余额 2026-09-23 15:26:44 +08:00
alfadb ae2735cdb6 feat(opencode): 用量窗口支持同 Key 组共享、活动防抖与超窗强刷
与 Ollama Cloud usage 对齐:OpenCode Go 作为同 Key 聚合订阅,
多个 openai/apikey/base_url=opencode.ai/zen/go/v1 账号共享同一份
用量状态与刷新节奏:

- 组共享:组指纹 sha256("opencode.ai\0"+api_key);auto_refresh 开关
  与 snapshot 按组写(事务 + FOR NO KEY UPDATE + 锚点 CAS + 纯合并,
  写 snapshot 不会抹掉开关);列表/详情经 ResolveAccounts 组内共享;
  账号换 Key → IdentityChanged → 组级关闭 auto_refresh 并清快照
- 活动防抖 + 超窗强刷:settings 新增 debounce_minutes(默认 1,
  1-60,必须小于 interval_minutes);模型请求活动经网关 5 处调用点
  驱动刷新,due = min(lastUsed+debounce, fetchedAt+maxWait),
  成功路径有 5 分钟最小抓取间隔,失败路径退避优先
- 手动刷新限频 10s→30s 且按组;singleflight/RunDue 按组去重;
  ListDue 在 SQL 内按组算 due(CTE 分组取组内 MAX(last_used_at))

验证:go build + 五包回归 OK;新增组共享/三态 due/组单飞/
IdentityChanged/debounce 校验单测全绿;vitest 11 例全绿;
vue-tsc 零错误。
2026-09-23 15:26:35 +08:00
alfadb 6624947b62 feat(opencode): 支持 OpenCode Go 官方用量窗口查询与自动刷新
OpenCode Go 上游(https://opencode.ai/zen/go/v1)官方提供用量端点
GET /zen/go/v1/usage(Bearer 复用消费 key),返回 rolling(5h)/weekly/monthly
三窗口的 percent 与 resetsAt。本提交为 openai apikey 账号(base_url 为
opencode.ai/zen/go/v1)接入该用量窗口:

- 后端:OpenCodeGoUsageService——手动刷新(每账号 10s 节流)、
  周期自动刷新(全局开关 + 账号开关双开启,interval 5-1440 分钟,
  leader lock + 每轮 20 账号 + 并发 4)、失败指数退避(上限 24h)、
  Retry-After 尊重;401→unauthorized、403→failed;snapshot 持久化于
  accounts.extra(单账号写入,无分组 CAS、无会话加密)
- admin API:settings GET/PUT、/:id/opencode-go-usage GET、
  auto-refresh PUT、refresh POST(与 Ollama Cloud usage 同构)
- 前端:账号列表行内三窗口进度条(复用 UsageProgressBar)、
  编辑弹窗内配置面板(详情+手动刷新+auto-refresh)、全局设置卡片、
  en/zh i18n

验证:
- 上游端点已用生产真实 Go key 实测 200,响应
  {"usage":{"rolling":{"status":"ok","percent":6,...},...}}(2026-08-13)
- go build ./... OK;service/handler/repo/dto/cmd 全量回归 OK
- 新增单测 20 例(解析 401/403/畸形 JSON、退避、节流等)全绿
- frontend vitest 新增 7 例 + Ollama 回归 4 例全绿;vue-tsc 零错误
2026-09-23 15:26:35 +08:00
Wesley Liddick afd069b169 Merge pull request #7532 from alfadb/feature/claude-cli-version-sync
feat(claude): 自动同步 Claude Code 客户端版本号,无需改代码跟随官方版本
2026-09-23 14:54:36 +08:00
alfadb ca08825935 feat(claude): 自动同步 Claude Code 客户端版本号
出站伪装用的 Claude Code 版本号此前硬编码在 claude.CLICurrentVersion,
只能靠环境变量 SUB2API_CLAUDE_CLI_VERSION 覆盖且需重启;Anthropic 抬高
版本下限时(如 Opus 5.5 要求 >= 2.1.280)必须改代码发版才能跟上。

改为每小时从官方 GitHub Releases 同步最新稳定版,无需重启也无需发版。

- 新增 ClaudeCodeVersionSyncService:1 小时间隔,数据源 anthropics/claude-code
  的 releases/latest,失败回退最近 30 条列表;过滤 draft/prerelease;
  只向前推进,拉取失败保留上次值不降级;面板可关闭
- 新增三个设置项:claude_code_client_version(管理员手动固定)、
  claude_code_client_version_synced(同步服务独占写)、
  claude_code_version_auto_sync_enabled(开关,默认开)
- 取值优先级:面板手动值 -> 同步值 -> SUB2API_CLAUDE_CLI_VERSION -> 内置基线,
  每层经 IsSupportedCLIVersion 校验,含 60s 缓存与 singleflight
- 版本号运行期化:claude.DefaultHeaders 由包级 var 改为函数,新增
  EffectiveCLIVersion/DefaultUserAgent 与 SetCLIVersionResolver(未注入时
  与 CLIVersion() 等价);identity 的 defaultFingerprint 不再在包 init 固化,
  指纹下限与主版本超前校验改用运行期值
- 守住一致性铁律:buildUpstreamRequest 与 count_tokens 各取一次 UA 字符串,
  同时喂给出站 User-Agent 头与请求体 billing 的 cc_version,避免缓存翻转
  瞬间头体版本不一致被判非正版客户端
- 管理后台设置页新增配置区块,含中英文案与「当前同步到」只读展示
2026-09-23 13:35:44 +08:00
HuntercodeT ecba49f711 fix(gateway): let OpenAI passthrough accounts supplement model discovery instead of hiding mapped models
A single passthrough account made GetAvailableModels return nil for the whole
group, so every other account's model_mapping was dropped from /v1/models and
the Codex catalog. Treat passthrough accounts like unmapped ones: skip their
(possibly stale) mapping and let supplementUnmappedOpenAIModels add the
default set, while ordinary accounts' mappings still reach the list.

Refs #6647, #7517
2026-09-23 11:51:06 +08:00
Achord Chan b5b5f1f94b fix(model-plaza): apply independent video billing multipliers 2026-09-23 11:28:41 +08:00
lyy0709 c4c2e66070 feat(models): support GPT-6 Sol, Luna and Claude Opus 5.5 2026-09-23 11:27:49 +08:00