
Supermemory is a memory and continual-learning engine for AI agents, available via API, plugins and MCP, that extracts, stores and re-injects relevant context so any model can remember across sessions.
| 维度 / 指标 | ![]() | |
|---|---|---|
| 官方域名 | claude.ai | supermemory.ai |
| 定价模式 | 免费增值 | 免费增值 |
| 月访问量 | 12.53亿 (1,253,439,413) | 0 (0) |
| 月度环比流量变化 | +2.41亿 (+23.8%) | 0 (—) |
| 主要市场与份额 | 🇺🇸 美国 (24.5%) | 🌐 全球 (—) |
| 直接访问 / 月 | 9.03亿 | 0 |
| 用户评分 | 4.8 | 4.4 |
| 支持的平台 | 网站 | 浏览器扩展 |
Anthropic 的前沿 AI 工作空间,由 Claude 5.5 驱动,擅长智能体式编程、分析与细腻写作。
结合前沿的 Claude Opus 5.5 与高速的 Sonnet 5.5,支持可配置的扩展思考、最高 100 万 token 的上下文窗口与 128K token 输出。
Pro、Max 与 Team Premium 方案打包了 Claude Code,可自主执行命令行软件工程、运行测试并在整个代码库范围内重构。
在聊天旁的专属工作空间面板中渲染可交互的实时 Web 应用、可视化图表、电子表格与排版好的 PowerPoint 演示文稿。
让对话基于共享的项目知识库,并通过 MCP 直接连接本地文件系统、GitHub、Slack 与内部数据库。

Supermemory is a memory and continual-learning engine for AI agents, available via API, plugins and MCP, that extracts, stores and re-injects relevant context so any model can remember across sessions.
A dedicated learner model extracts, merges, infers and forgets facts, storing each learning in a vector-graph database that understands entities, sources and time.
Learnings are injected into the model's context in real time via hooks, so agents recall relevant information without stuffing full histories into the prompt.
Memory lives outside any single model provider, so it carries across models and harnesses and is delivered through the API, plugins and MCP.
Mandatory memory is only billed once: repeated or unchanged content is not charged again, and the same rates apply on every plan.

拥有超大上下文窗口与深度的 Google Workspace 集成,但在细腻的创意写作和代码掌控上稍逊一筹。
拥有超大上下文窗口与深度的 Google Workspace 集成,但在细腻的创意写作和代码掌控上稍逊一筹。
免费 / 开源API 费用显著降低且具备强大的开源推理模型,但消费级聊天界面的精致度稍弱。

Real-time access to live social media data and rapid image generation, but less reliable for technical coding tasks.
专项更强Fully autonomous agent executing complex multi-step web tasks, but less suited for conversational brainstorming and iterative writing.

免费 / 开源Fully open-source multi-model interface connecting your own API keys, but requires manual setup and hosting.