I'm having difficulty finding information on this. Much like Astra and Claude can make use of your computer through their desktop apps and handle tasks for you in that way... does Odysseus have computer use capabilities as well? Assuming you're running an LLM that also supports computer use of course.
I'm new to Hermes but I was using it just fine for 2 days straight. Using Nex N2 through OpenRouter and talking to it via Telegram. But this morning I give it even basic tasks like renaming files and it's just timing out. Has anyone see this problem before? Any thoughts on how to fix it?
Hi, I installed Hermes via command line on Mac yesterday and I have it set up using Nex-N2 via OpenRouter, with Telegram as my means to chat with it. Some quick questions here: I've noticed that when I tell it to do things it re-processes older messages sometimes in its thinking which spends a lot of time and theoretically a lot of tokens, but thank goodness Nex-N2 is still free. What's the best way to get it to stop doing this and to stop having it repeat old answers at the beginning of new ones?
Second, if I eventually decide to change the AI model powering it to something like Claude or OpenAI or Kimi... will it still remember things ok? I know it's writing to a memory file somewherr but is there any real negative impact to trying out different models periodically and seeing which I like? Similarly would it be a bad idea to have it help me code something and then swap out models mid- project?
Finally what are you using it with? What's been everyone's preferred setup for your specific tasks or needs?
