能学习的 AI Agent。
Hermes Agent 是一款开源 AI Agent，内置学习闭环——从实践中积累技能、跨会话保留记忆，并在你选择的任意模型或平台上运行。
为思考、学习和成长而生
闭环学习系统
完成复杂任务后，Hermes 会自动创建技能并在使用中持续优化。它主动将知识沉淀为可复用的技能，并能检索历史对话——每次会话都比上次更聪明。
随你所在
通过单一网关进程接入 Telegram、Discord、Slack、WhatsApp 或 Signal。支持语音备忘录转写和跨平台对话连续性——随时随地都能触达你的 Agent。
不只在你的笔记本上运行
六种终端后端——本地、Docker、SSH、Daytona、Singularity 和 Modal。部署在 $5 的 VPS 或 GPU 集群上均可。无服务器持久化意味着空闲时几乎零成本。
不绑定任何模型
一条命令即可在 Nous Portal、OpenRouter、OpenAI、Kimi、MiniMax、GLM 或自定义端点之间自由切换。200+ 个模型可选，无需改动代码，零迁移成本。
三条命令，搞定一切。
安装
支持 Linux、macOS、WSL2 和 Termux（Android）。
配置
配置模型提供商、工具集和消息平台。
开始
开始你的第一次对话，Agent 已就绪。
支持任意模型
一条命令切换提供商，不锁定任何一家。
来自社区的声音
真实用户，真实工作流。
“I migrated from Openclaw -> Hermes and so far, so good — Things 'just work' a lot better — The transition of data from OC -> Hermes was very easy too — It doesn't seem to randomly crash and stop working on me — When I ask it to do things, it'll create an actual skill, rather than just relying or hoping it'll remember what I want next time.”
“i've been using @NousResearch Hermes Agent for about a week and my initial thoughts are it just works out of the box. I find it's memory and learning to be far superior to OpenClaw without augmenting it with QMD or any additional memory systems.”
“Just updated to Hermes Agent v0.4.0 and as a non-technical heavy user who lives in this thing every day… this one feels like it was made for people like me. No code, no hassle, just straight-up better.”
“I gave it feedback: 'Your relevance scores are too generous. Anything below 7 shouldn't make the cut.' It updated its approach — and because it has persistent memory, it remembered that calibration. Three weeks in, the digests are genuinely good. Not because I fine-tuned anything, but because the agent learned my taste through normal conversation.”
“I migrated from Openclaw -> Hermes and so far, so good — Things 'just work' a lot better — The transition of data from OC -> Hermes was very easy too — It doesn't seem to randomly crash and stop working on me — When I ask it to do things, it'll create an actual skill, rather than just relying or hoping it'll remember what I want next time.”
“i've been using @NousResearch Hermes Agent for about a week and my initial thoughts are it just works out of the box. I find it's memory and learning to be far superior to OpenClaw without augmenting it with QMD or any additional memory systems.”
“Just updated to Hermes Agent v0.4.0 and as a non-technical heavy user who lives in this thing every day… this one feels like it was made for people like me. No code, no hassle, just straight-up better.”
“I gave it feedback: 'Your relevance scores are too generous. Anything below 7 shouldn't make the cut.' It updated its approach — and because it has persistent memory, it remembered that calibration. Three weeks in, the digests are genuinely good. Not because I fine-tuned anything, but because the agent learned my taste through normal conversation.”
“it's pretty obvious at this point. Hermes Agent > OpenClaw”
“The core difference in one sentence: OpenClaw stores everything and searches it. Hermes keeps almost nothing in the prompt and retrieves the rest.”
“I am actually getting stuff done instead of debugging.”
“It has built-in learning — if something breaks, it ACTUALLY remembers it and creates a skill for troubleshooting it. So far, completely better experience.”
“Looking through code it looks like an actual app where openclaw is more like tech demo.”
“it's pretty obvious at this point. Hermes Agent > OpenClaw”
“The core difference in one sentence: OpenClaw stores everything and searches it. Hermes keeps almost nothing in the prompt and retrieves the rest.”
“I am actually getting stuff done instead of debugging.”
“It has built-in learning — if something breaks, it ACTUALLY remembers it and creates a skill for troubleshooting it. So far, completely better experience.”
“Looking through code it looks like an actual app where openclaw is more like tech demo.”
常见问题
~/.hermes/skills/准备好安装 Hermes Agent 了吗？
免费、开源，2 分钟内即可运行。