supermemory
@supermemoryai
Context engine for your LLMs, Personalized for your users. 15k+ total ⭐ on Github, 6 OSS projects. Join the community - https://supermemory.link/discord
Introducing: Infinite Chat API 💥 Extend the context length of ANY model ...while saving 90% of your tokens and cost AND improving performance! One single line to switch. Available to use now
Did you know? supermemory has built-in connectors to Google, Notion, Onedrive and others, which means it's really easy to sync data to and from these services! supermemory.ai/blog/building-…
one standard to let agents discover each other regardless of standards sounds familiar
After MCP, A2A, & AG-UI, there's another Agent protocol. It's fully open-source and launched by IBM Research. Here's a complete breakdown (with code):
These three are hands down the best built on/for LLM tools right now: @supermemoryai, @greptileai, and @coderabbitai. end-to-end experience across all of them is insanely well done. Just top tier products.
you can now view your memories in a more natural, related, and interconnected way more super cool stuff on the way ✌️
customers are loving graphs inside @supermemoryai You can now view how the supermemories of your users evolve, as documents and memories are added to it, with: temporal understanding, forgetfulness relevancy bias (..& more) literally just postgres btw. no graphRAG bullshit
We partnered with @PlanetScale to give you the most performant memory engine engine you've ever seen This means, for our customers supermemory is, - 40% faster - Cheaper - More reliable and performant (up to 1000 QPS!) supermemory.ai/blog/supermemo…
THE THINGS HAPPENING IN THIS SPACE ARE SO COOL (read more research)
"It's just RAG bro"
Guys..... this is just a fraction of the actual full graph for supermemory MCP 🤯
You can also now see a distribution of your top users by number of memories added!
supermemory MCP spotted 🥺
how tf do you remove problem mcps from claude code?
BEHOLD: Supermemory's graph You can now view the entire constructed supermemory graph on the supermemory console. You can view the documents, memories extracted out, and more information about the links between the memories, and how it's self-forming and self-evolving ✨
I heard you like graphs?
We @supermemoryai switched to @DocmostHQ Delightful UX Built for collaboration with really good access controls Completely self-hostable. Paid license is cheaper than all alternatives Privacy focused And open-source with AGPL license 🤯
Product idea: Notion except every keystroke doesn't feel like I'm SSH'd into a server on Mars.
This is why it's so important to engineer your context to be able to specifically answer queries LLM context windows is still an unsolved problem. we call this "Context pollution", or "rot" - In our own benchmarks @supermemoryai, building on top of the NoLiMa paper of Adobe,…
Introducing our latest technical report: Context Rot - How Increasing Input Tokens Impacts LLM Performance Our results reveal that models do not use their context uniformly. full report in replies
For everyone looking - mcp.supermemory.ai
💡 Startup idea: I’ve trained ChatGPT to know me. My style, my preferences, my work. But if I want to use Grok or Claude, I have to start from scratch. There should be a private memory that I own and control that moves with me from one AI to another. An API of me.
amazing idea! you should check out supermemory, they are doing the same thing :)
💡 Startup idea: I’ve trained ChatGPT to know me. My style, my preferences, my work. But if I want to use Grok or Claude, I have to start from scratch. There should be a private memory that I own and control that moves with me from one AI to another. An API of me.
<有趣的GitHub项目：supermemory-mcp> 让你的记忆不再局限于ChatGPT，Universal Memory MCP让所有语言模型都能访问你的记忆。无需登录、完全免费，一条命令即可设置。🚀 极速且可扩展，简单易用！ github.com/supermemoryai/…
this guy is the perfect example of founder mode and the absolute goat of devx if you need conversational memory for llms please check out @supermemoryai
no worries bro, memory is my badsignal. if we're shitty, i would love to be told that we're shitty, our rate of improvement depends on the amount of criticism we get so that's what I live and die for, please complain
Knowledge graphs are a pain to work with, so here's a starter for beginners - supermemory.ai/blog/knowledge…
