用户协议与隐私政策

本页包含《用户协议》与《隐私政策》,适用于蔷薇智脑 Token 平台。

本协议是您与强魏飚唐(上海)智能科技有限公司(下称「我们」)就使用蔷薇智脑 Token 平台(下称「本平台」,包括其网站、AI 网关接口与在线工作台)所订立的条款。注册账号或以任何方式使用本平台服务,即视为您已阅读、理解并同意本协议的全部内容。

最近更新:2026 年 10 月 3 日

第一条定义

1.1API Key:注册完成后系统自动生成的接口调用凭证,以您的团队身份发起调用与计费。

1.2访问令牌(PAT):您为智能体或脚本签发的访问凭证,可设定有效期,并随时停用或删除。

1.3用量:模型调用所消耗的输入与输出 token 数量,按平台公示的计价规则折算为费用。

第二条账号与租户

2.1注册需提供团队或个人名称、登录账号与密码;您应确保信息真实、准确,因信息不实造成的后果由您承担。

2.2注册即创建团队租户,租户内的余额、API Key 与用量统一核算。

2.3您应妥善保管账号密码与各类凭证;凭证泄露造成的冒用损失由您自行承担,发现泄露应立即停用相关 Key 或令牌并联系我们。

第三条服务内容

3.1本平台提供 OpenAI 兼容的模型推理网关、用量计量与余额管理、API Key 与访问令牌管理等服务,具体以平台页面实际提供的为准。

3.2可用模型范围、上下文长度限制与并发配额以平台公示为准,我们可能根据上游能力进行调整。

3.3模型输出由人工智能生成,可能存在不准确之处;您应在生产环境中自行核验后再行使用。

第四条计费与充值

4.1平台采用预充值模式,余额不足以支付当次调用时,调用将被拒绝。

4.2充值经微信支付完成,到账以支付渠道回执为准。

4.3计价规则在平台内公示;规则调整前于平台内公告,调整不影响已完成的调用。

第五条使用规范

您承诺不以任何方式:

  • 利用本平台从事违法犯罪活动,或生成、传播违法内容;
  • 侵害他人知识产权、商业秘密、个人信息或其他合法权益;
  • 对平台进行反向工程、漏洞扫描、压力测试或其他恶意攻击;
  • 将接口转售或提供给未获授权的第三方使用,或以技术手段规避计量与限流。

5.5违反上述约定的,我们有权暂停或终止服务,并保留追究法律责任的权利。

第六条服务可用性

6.1因维护、升级导致的服务中断,我们将提前在平台内公告,紧急情形除外。

6.2上游模型服务异常可能影响网关调用,我们将尽力协调恢复,但不承诺无条件可用。

6.3我们可能新增、调整或下线部分功能;重大变更在平台内公告。

第七条知识产权与数据

7.1本平台的软件、界面、文档及相关知识产权归我们或相应权利人所有,未经许可不得复制或商用。

7.2您上传的数据与调用产出的内容归您所有;除计量计费、安全风控与故障排查所必需外,我们不会访问您的业务数据。

第八条免责与责任限额

8.1服务按「现状」提供,我们在法律允许范围内不对服务的连续性、及时性与适配性作明示或默示担保。

8.2对因使用或无法使用本平台导致的利润损失、数据损失等间接损失,我们不承担责任;我们对您的累计赔偿责任,以您最近十二个月内的实际充值总额为限。

第九条协议变更与终止

9.1我们可对本协议进行修订并在平台内公告;公告后您继续使用服务的,视为接受修订。

9.2您可随时停止使用;注销账号请联系我们办理,注销后您的数据将依法删除或匿名化,法律法规另有规定的除外。

第十条法律适用与争议解决

10.1本协议的订立、履行与解释适用中华人民共和国法律。

10.2因本协议产生的争议,双方应友好协商;协商不成的,提交上海市有管辖权的人民法院解决。

These Terms of Service (the "Agreement") govern your use of the 蔷薇智脑 Token Platform (the "Platform", including its website, AI gateway APIs and online workbenches) and constitute the agreement between you and 强魏飚唐 (Shanghai) Intelligent Technology Co., Ltd. ("we", "us"). By registering an account or using the Platform in any way, you acknowledge that you have read, understood and accepted all provisions of this Agreement.

Last updated: October 3, 2026

Article 1Definitions

1.1API Key: the interface credential automatically generated upon registration, used to place calls and incur charges under your team's identity.

1.2Access Token (PAT): a credential you issue for agents or scripts, with a configurable lifetime, which can be disabled or deleted at any time.

1.3Usage: the number of input and output tokens consumed by model calls, converted into charges per the pricing rules published on the Platform.

Article 2Account and Tenant

2.1Registration requires a team or individual name, a login account and a password. You warrant that the information is true and accurate, and you bear the consequences of any inaccuracy.

2.2Registration creates a team tenant; the balance, API Keys and usage within a tenant are settled together.

2.3You must keep your credentials secure. Losses caused by leaked credentials are borne by you; upon discovering a leak, immediately disable the relevant Key or token and contact us.

Article 3Services

3.1The Platform provides an OpenAI-compatible model inference gateway, usage metering and balance management, and API Key / access token management, as actually presented on the Platform.

3.2The range of available models, context length limits and concurrency quotas follow what is published on the Platform and may change with upstream capacity.

3.3Model outputs are generated by artificial intelligence and may be inaccurate; verify them before using them in production.

Article 4Billing and Top-up

4.1The Platform uses a prepaid model; calls are rejected when the balance cannot cover them.

4.2Top-ups are processed via WeChat Pay; crediting is subject to the payment channel's receipt.

4.3Pricing rules are published on the Platform; changes are announced in advance and do not affect completed calls.

Article 5Acceptable Use

You agree not to:

  • use the Platform for illegal activities, or to generate or disseminate unlawful content;
  • infringe intellectual property, trade secrets, personal information or other lawful rights of any third party;
  • reverse engineer, vulnerability-scan, stress-test or otherwise attack the Platform;
  • resell or provide the APIs to unauthorized third parties, or evade metering and rate limits by technical means.

5.5If you breach the above, we may suspend or terminate the service and reserve the right to pursue legal liability.

Article 6Availability

6.1Service interruptions caused by maintenance or upgrades are announced on the Platform in advance, except in emergencies.

6.2Upstream model outages may affect gateway calls; we will coordinate recovery but do not promise unconditional availability.

6.3We may add, adjust or retire features; material changes are announced on the Platform.

Article 7Intellectual Property and Data

7.1The software, interfaces, documentation and related intellectual property of the Platform belong to us or the respective rights holders, and may not be copied or commercialized without permission.

7.2Your uploaded data and the content produced by your calls belong to you; except as necessary for metering, billing, security and troubleshooting, we do not access your business data.

Article 8Disclaimer and Limitation of Liability

8.1The service is provided "as is"; to the extent permitted by law, we make no express or implied warranty of continuity, timeliness or fitness.

8.2We are not liable for indirect losses such as lost profits or data loss arising from the use of or inability to use the Platform; our aggregate liability to you is limited to your actual top-up total in the preceding twelve months.

Article 9Changes and Termination

9.1We may revise this Agreement and announce the revision on the Platform; continued use after the announcement constitutes acceptance.

9.2You may stop using the Platform at any time; contact us to cancel your account. After cancellation, your data will be deleted or anonymized as required by law.

Article 10Governing Law

10.1This Agreement is governed by the laws of the People's Republic of China.

10.2Disputes shall first be resolved through negotiation; failing that, they shall be submitted to the competent people's court in Shanghai.

本政策说明蔷薇智脑 Token 平台收集哪些信息、如何使用与保护这些信息。我们按「最小必要」原则收集数据,不出售您的个人信息。

最近更新:2026 年 10 月 3 日

一我们收集的信息

1.注册信息:团队或个人名称、登录账号。密码经单向加密存储,任何人(包括我们)均无法还原明文。

2.调用记录:每次模型调用的接口、模型名称、输入与输出 token 数量及时间,用于计量计费。

3.凭证信息:API Key 与访问令牌的名称、前缀、状态与有效期;密钥本体加密存储,仅创建时明文展示一次。

4.订单信息:充值订单号与金额。支付经微信支付完成,我们不存储银行卡等敏感支付数据。

5.日志信息:访问时间、IP 地址与浏览器类型,用于安全审计与故障排查。

二信息的使用

我们仅为以下目的处理您的信息:提供服务与计费、账号与凭证管理、安全与反滥用、服务改进与故障排查。除本政策载明的情形外,我们不会将您的信息用于其他目的。

三本地存储说明

登录状态、语言偏好以及您在智能标注工作台中填写的 API Key,均保存在您浏览器的本地存储(localStorage)中,不上传至我们的服务器。清除浏览器站点数据即可删除上述内容。

四存储与保护

您的数据存储于中华人民共和国境内的服务器,传输过程采用 HTTPS 加密。我们采取访问控制、日志审计等措施防止数据被未经授权地访问、篡改或泄露。数据仅在实现本政策所述目的所必需的期限内保留。

五信息的共享

除以下情形外,我们不会向任何第三方提供您的个人信息:

  • 取得您的明确同意;
  • 为完成充值,与微信支付之间进行的必要信息交互(仅限订单必需信息);
  • 法律法规或主管部门的强制性要求。

六您的权利

您有权查询、复制、更正与删除您的个人信息,亦可注销账号。对您提出的相关请求,我们将在核实身份后于合理期限内处理。注销账号后,我们将依法删除或匿名化您的数据,法律法规要求留存的除外。

七未成年人

本平台面向企业级用户,不接受未成年人注册与使用。

八政策更新

本政策如有重大更新,我们将在平台内公告。继续使用即视为您知悉并同意更新后的政策。

九联系我们

本政策由强魏飚唐(上海)智能科技有限公司发布。如有疑问或需行使上述权利,请通过官网「联系我们」与我们取得联系,我们会尽快回复。

This policy explains what information the 蔷薇智脑 Token Platform collects and how it is used and protected. We collect the minimum data necessary and never sell your personal information.

Last updated: October 3, 2026

1Information We Collect

1.Registration information: your team or individual name and login account. Passwords are stored with one-way encryption and cannot be recovered in plain text by anyone, including us.

2.Call records: the API, model name, input and output token counts and timestamps of each model call, used for metering and billing.

3.Credential information: the name, prefix, status and expiry of your API Keys and access tokens; the secrets themselves are encrypted and shown in plain text only once at creation.

4.Order information: top-up order numbers and amounts. Payments are processed by WeChat Pay; we do not store sensitive payment data such as bank card numbers.

5.Log information: access time, IP address and browser type, used for security auditing and troubleshooting.

2How We Use Information

We process your information only to deliver and bill the service, manage accounts and credentials, ensure security and prevent abuse, and improve the service and troubleshoot. Beyond the scenarios described in this policy, we do not use your information for other purposes.

3Local Storage

Your sign-in state, language preference and the API Key you enter in the smart annotation workbench are stored in your browser's local storage (localStorage) and are not uploaded to our servers. Clearing site data in your browser removes them.

4Storage and Protection

Your data is stored on servers located within the People's Republic of China and transmitted over HTTPS. We apply access controls and log auditing to prevent unauthorized access, tampering or disclosure. Data is retained only as long as necessary for the purposes described in this policy.

5Sharing

We do not provide your personal information to third parties except:

  • with your explicit consent;
  • to exchange the information necessary to complete a top-up with WeChat Pay (limited to order-essential information);
  • as required by laws, regulations or competent authorities.

6Your Rights

You may access, copy, correct and delete your personal information, and may cancel your account. We process verified requests within a reasonable period. After cancellation, your data will be deleted or anonymized as required by law, except where retention is mandated.

7Minors

The Platform serves enterprise users and is not open to minors.

8Updates to This Policy

Material updates to this policy are announced on the Platform. Continued use means you understand and accept the updated policy.

9Contact Us

This policy is published by 强魏飚唐 (Shanghai) Intelligent Technology Co., Ltd. For questions or to exercise the rights above, reach us through "Contact Us" on the website and we will respond promptly.