OpenRouter(@OpenRouterAI)
Cache is scoped per API key, so different keys under the same account stay isolated. Cache hits don'...
7.2内容质量
TL;DR · AI 摘要
OpenRouter 推出基于 API Key 隔离的响应缓存功能，缓存命中不消耗下游模型提供商的速率限制，目前处于 Beta 阶段。
核心要点
- 缓存作用域严格限定在单个 API Key 内，同账户下多 Key 互不干扰
- 缓存命中请求不会转发至后端模型提供商，因此不计入其速率配额
- 该功能已上线 Beta 版，文档地址已公开
结构提纲
按章节快速跳转。
OpenRouter 宣布推出响应缓存功能，当前处于 Beta 阶段。
缓存按 API Key 粒度隔离，同一账户下的不同 Key 无法共享缓存。
缓存命中请求不经过下游模型提供商，因此不占用其速率配额。
适用于高频重复请求、调试与灰度验证等低延迟高复用场景。
思维导图
用一张图看清主题之间的关系。
查看大纲文本（无障碍 / 无 JS 友好）
- OpenRouter 响应缓存
- 作用域
- 按 API Key 隔离
- 限流影响
- 缓存命中不计入下游限频
- 状态与文档
- Beta 中，含官方文档链接
金句 / Highlights
值得收藏与分享的关键句。
Cache is scoped per API key, so different keys under the same account stay isolated.
Cache hits don't count against provider rate limits since the request never reaches them.
In beta now. Docs: https://openrouter.ai/docs/guides/features/response-caching…
#OpenRouter#LLM#API#缓存#速率限制
打开原文In beta now. Docs: https://t.co/BqnCYIFvbg" / X
OpenRouter on X: "Cache is scoped per API key, so different keys under the same account stay isolated. Cache hits don't count against provider rate limits since the request never reaches them. In beta now. Docs: https://t.co/BqnCYIFvbg" / X
Don’t miss what’s happening
Cache is scoped per API key, so different keys under the same account stay isolated. Cache hits don't count against provider rate limits since the request never reaches them. In beta now. Docs: https://openrouter.ai/docs/guides/fe atures/response-caching…
·
2
18
6