@Teknium @teortaxesTex @max_paperclips Quick question. Which are the differences in running hermes or hermes—Tui?
English
Joan Subirats
3.7K posts
Joan Subirats
@Subirats345
Cuenta Oficial de Joan Subirats.
Barcelona Katılım Eylül 2010
718 Takip Edilen409 Takipçiler
Comparing deepseek-tui, opencode and Hermes side by side with the same tasks to V4-Pro, I am fairly confident that Hermes is the best agent among them right now. Highest success rate, fastest, and cheapest. A bit of a shame because the other two appeal to me aesthetically.
English
@petergyang The mistery for me is why it works lot better on Hermes! I tried just for curiosity and it’s like night and day. Why @NousResearch ?
English
I know OpenClaw isn't part of OpenAI but this feels like a mini-crisis for OpenAI if the GPT integration doesn't improve soon.
The bar is GPT needs to be just as good if not better for OpenClaw as Opus.
English
@Teknium @chrisbward so SSH seems like a partial answer. I think the remaining distinction is conceptual: remote execution is one thing, but a node model is closer to “this same agent can operate through this device’s own capabilities” rather than just opening a shell there. @NousResearch
English
@Teknium @chrisbward my concern was never just “how do I run commands on another machine?”. It was more: how do I avoid ending up with multiple Hermes installs that feel like multiple separate agents, each with their own memory/config/state? 👇
English
Even anthropic is copying from us now, we must be doing something right!
3 days ago we added notify when done, so the agent does not have to poll the background process(es) that it spawns, and instead when it finishes it can queue up a new message for the agent. Work on other things in the same session while it works!
Open Source can move faster than any closed centralized company.
See the PR here: github.com/NousResearch/h…
Noah Zweben@noahzweben
Thrilled to announce the Monitor tool which lets Claude create background scripts that wake the agent up when needed. Big token saver and great way to move away from polling in the agent loop Claude can now: * Follow logs for errors * Poll PRs via script * and more!
English
@Teknium @chrisbward what I like in the gateway + nodes OpenClaw's approach is the idea of one agent identity that expands across devices. The important part is not only execution, but ATTACHED capabilities: local tools, local context, device-native integrations, etc.
English
@Teknium @chrisbward Thanks, really appreciate the reply :)
SSH backend definitely helps, because it means I could keep one Hermes running 24/7 on the VPS and still have it execute things on my Mac. but... 👇
English
@Teknium @chrisbward Hermes feels so stable and clean imo, I'm enjoying it a lot! But it's like trapped in the VPS, and if I install it too on another device, I feel that there are 2 complete different entities
docs.openclaw.ai/nodes
English
@Teknium @chrisbward This works pretty well in OC, where we can have a single gateway and pair nodes (devices) to to it. All the messages reaches the gateway, but he can perform things and can have access in the nodes. So one single instance, memory, config, etc... growing and expanding
English
@Teknium @chrisbward The gateway + paired nodes strategy is important imo. If we install Hermes on every host we want to exec actions, then all the configs, sessions, skills, memories, etc... will differ right? In every one of those machines, we'll be talking to a different agent, different state...
English
@chrisbward Still havent read the docs but that was just my first thought.
i.e. install hermes on rpi. Set SSH as terminal backend, all code would execute on that machine, it's filesystem would use that machine etc.
But will try to update once Ive had a chance to look over their docs!
English
Excellent recommendations yesterday for places to work in Tokyo. Ended up at Share Lounge marunouchi and it’s exactly what I was looking for 👍
English
De plato, pollo de corral a la crema, al más puro estilo de Bocuse.
Español