EverMind @evermind

拥有 Harness of harnesses，以及 RSI （递归自我进化）架构的 Raven，国庆节也要休息休息啦！ 欢迎大家持续关注 @evermind 最新 harness 架构的更多动态哦！目前我们的技术报告也已经在 arxiv 上发布啦！欢迎大家查看！

七仔 @Yangtze_Seventh

5 days ago

模型一直在升级，是不是可以不需要给 「Agents」当项目经理了？ 前几天我给大家分享了 Raven 做的游戏、PPT，还有调研和实验案例，在不同程度上说明了现在多 Agent 的能力边界，无论是视觉效果还是功能，都很强大。 现在让 AI

keyboard_arrow_left Previous keyboard_arrow_right Next

2 0 9 32K 3

Raven 🐦‍⬛ reached #1 on Hugging Face Papers’ weekly list. We’ve spent a lot of time working on how specialist agents coordinate and carry useful experience into their next task. Seeing people pick up the paper means a lot to the team. Thanks to everyone who read it, upvoted, or tried the code. Raven is open source. Give it a run. We’d love to hear what worked, what broke, and what you’d build with it. github.com/EverMind-AI/Ra…

@LongTermMemoryE 🐦‍⬛ 🏆

Check out this Raven 🐦‍⬛ skill. Raven created this video under direction, and the entire workflow is now available as a reusable skill in the Raven repository.

海拉鲁编程客 @hylarucoder

7 days ago

已开源 这可能是市面上为数不多，写给技术人员的非技术 Skill 如有帮助还请点赞转发支持一下！ x.com/i/article/2104…

22 28 219 108K 323

EverOS is now a @convex-dev component. 🧠 Add persistent long-term memory to any Convex app, or to @convex-dev/agent in one line, backed by EverOS Cloud. Convex already gives agents the backend primitives they need: a reactive database, durable functions, scheduling, and threads. Now agents can remember across threads and sessions too. EverOS turns conversations into durable user memory: facts with timestamps and traceable sources, plus a profile that evolves over time. That memory belongs to the user and follows them across threads, sessions, models, and agents. Under the hood, EverOS delivers state-of-the-art results across long-term memory benchmarks. No vector database to run. No memory infrastructure to stitch together. Get it today: convex.dev/components/eve…

用 Opus 5.5 搓了个产品故事视频！ 去年 7 月想做类似的事，受限于大模型的能力无法完成，没想到现在就进步到这个程度了！ 如果是去年的我看到这个，直接晕眩瘫软仿佛看到原子弹爆炸。 BTW: 本视频由开源版 Raven v0.2.0 指挥 Claude Code 生成 如果点赞多我会整理成 skill 放出来！@evermind

Two months ago, we, @evermind, launched Raven 🐦‍⬛. The next version is almost here, and I think this game shows where we're headed better than a diagram could. We built this 3D boss fight with Raven. Watch the lighting, the two boss phases, the weapons, and the ending. Please turn the sound on too. I'm genuinely impressed by how it all came together. It took real compute and plenty of iteration. What excites me is Raven coordinating specialized agents on one task graph to make something that feels like one game. That's what we mean by “Harness of Harnesses.” More soon. Raven is open source: github.com/EverMind-AI/Ra…

@dinah_zhang @elliotchen We had a feeling yellow was your color. 💛

Running several agents gets messy fast. Raven is built to coordinate them. Raven is a pre-alpha multi-agent orchestration repo for builders who want to coordinate specialized agents from one place. It helps you delegate work across research, coding, design, and unattended workflows by bringing built-in and third-party agents into shared workflows. Key features: • Unified orchestration surface – delegate tasks, coordinate execution, and integrate results in one place • Four built-in agents – Raven-Research, Raven-Code, Raven-Design, and Raven-Oncall cover distinct workflow types • Deep research support – Raven-Research produces structured reports with traceable sources for complex questions and technical analysis • Agentic development workflows – Raven-Code supports implementation, debugging, refactoring, data processing, and analysis • Third-party agent connections – connect external agents and coordinate their capabilities in shared workflows It’s open-source under the Apache License 2.0. Link in the reply 👇

The more I look into @evermind, the more I think the real problem they’re trying to solve isn’t intelligence. It’s 𝐜𝐨𝐧𝐭𝐢𝐧𝐮𝐢𝐭𝐲. AI agents can already write code, use tools, call APIs and complete multi-step tasks. But once the session ends, a lot of that context disappears. @evermind is building the infrastructure for agents to carry that experience forward. 🧵🔻

keyboard_arrow_left Previous keyboard_arrow_right Next

Meet EverMe for Chrome. Turn your ChatGPT, Claude, and Gemini conversations into memory you can keep and reuse. The preferences you’ve shared. The projects you’re working on. The context you’ve spent time explaining. Keep chatting as usual. EverMe syncs completed conversations to your Memory Hub, where you can review the extracted memories and use them across EverMe-supported agents. Already have a history worth keeping? Import your past ChatGPT and Claude conversations, too. Choose which platforms to sync. See what’s captured. Keep your memory in your hands. chromewebstore.google.com/detail/akabimn…

keyboard_arrow_left Previous keyboard_arrow_right Next

@Shashwat_web3 @Letta_AI @ShunyuYao12 @wujiang_ai @zep_ai I prefer EverOS. 💛

.@milvusio is supported in EverOS 👀

EverMind @evermind

4 weeks ago

EverOS 1.3.0 is out. You can now use @milvusio or Zilliz Cloud to index your agent’s memory. LanceDB remains the default. Markdown remains the source of truth. Switching backends means rebuilding the index from your memory files. Already running Milvus? EverOS now fits into

7 0 16 22K 2

盛大集团带着百万级投资来找项目了，正在做产品的朋友看过来👇 NovaCatapult「超级个体弹射器」正式招募开发者和创业团队，人民币 100 万起投，帮你把想法做成产品，也让好产品有机会被更多人看到。 除了资金，早期创业容易缺的几块支持也是全乎的： 算力：内部集群优先调用，减轻采购 GPU 的压力（做算法的都知道有多少含金量）。 技术：资深工程师帮你评审架构，遇到问题有人分析拆解，提出合理方案。 增长：增长团队协助获客，并有机会对接 @evermind @TankaChat 、@thetawellnessai、@makeplayai、@alayastd、@大圆镜 等生态渠道。 后台：法务、HR、财税等支持，让团队少为产品之外的事情分心。 欢迎独立开发者和小团队，尤其是已经动手、有想法有 Demo，准备全职把项目做下去的朋友。 补充一些小 tips：「超级个体弹射器」是投资孵化，不是无偿补贴；百万方案包含 50 万现金＋50 万算力等资源，涉及股权及后续转换条款；核心创始人需要全职到岗，目前以上海、北京为主，具体大家可以到官网查看～ 感兴趣可以直接官网报名，也欢迎 DM 我，发一段项目介绍、目前进展和 Demo 链接。我可以通过 @evermind帮大家做内部推荐，对接后续评审。 期待大家天马行空的作品！

Our CEO, Yafeng Deng, joined INCLUSION · Conference on the Bund to share EverMind’s work on memory and agent self-evolution. The talk explored the whole agent, rather than the LLM alone, as a possible analogy for the human brain, and the prospect of building a trainable agent framework that continuously learns from user data. Memory is central to that exploration. Thank you to everyone who came in person and asked thoughtful questions. We appreciated the chance to discuss the work face to face.

keyboard_arrow_left Previous keyboard_arrow_right Next

From 8,594 tokens to 1,946 per answer. EverOS selects the memories a question actually needs instead of injecting a fixed 20 every time. In our LongMemEval-S test, that meant 77% fewer answer-stage input tokens, with 91% accuracy versus 93.4%. Less context to process. Relevant evidence to work with. github.com/EverMind-AI/Ev…

EverOS 1.3.1 is here. The multi-round retrieval we just shared is now available in EverOS: let the model decide what evidence to keep, what to search next, and when to stop. Also in this release: • One reproducible runner for four memory benchmarks: LoCoMo, LongMemEval, EverMemBench, and SubtleMemory • A separately configurable retrieval decider • Better timeout handling, retry behavior, and database diagnostics • No storage migration or index rebuild required. Multi-round retrieval is opt-in. github.com/EverMind-AI/Ev…

How many memories does an answer actually need? One query needs a single fact. Another needs several pieces of evidence. A fixed top-k gives both the same budget. EverOS uses multi-round retrieval: a small model reads candidate episodes, selects the core evidence, and decides whether to stop or search again for what’s missing. In our LongMemEval-S injection ablation, core-only averaged 1.95 episodes per query: - 77% fewer answer-stage prompt tokens than k=20 - 91.0% accuracy, versus 93.4% at k=20 Retrieve again when evidence is missing. Stop when there’s enough. Let the query determine the memory budget. evermind.ai/blogs/multi-ro…

SkillCorpus just crossed 600 GitHub stars, and our demo dataset passed 100 downloads on Hugging Face in the past month. Thanks for taking a look. If you’re new here: SkillCorpus helps agents find procedures for the task in front of them. It collects public SKILL.md files, applies quality, safety, and license checks, removes duplicates, and retrieves relevant instructions, references, and scripts. Recent updates include OpenClaw 2.0 support and on-demand skill_search. Your agent can look up a procedure partway through a task, when it needs one. We also support retrieval across local skills, EverMind SkillHub, ClawHub, and skillhub.cn, with filtering and deduplication before the final selection. Want to explore the pieces yourself? The curation pipeline, embedding model, reranker, and a 1,000-skill sample are public. The sample includes the supporting files shipped with those skills. More skills are coming. What’s a task your agent still needs too much hand-holding to finish? github.com/EverMind-AI/Sk…