Rendered at 20:49:01 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
antonios_makro 11 hours ago [-]
This is exactly the right direction. A native desktop client is what turns these models from a tech demo into an actual tool. We've been focused on the other side of that problem: giving local models a face and a voice. We open-sourced our work as NyxClaw.ai, it will be a fun component to pair with this.
goran-j 11 hours ago [-]
Microsoft was showcasing OpenClaw desktop app in they keynote @ Build conference, it would be interesting to see how it compares to Hermes desktop and will it be extensible so it can be used with NyxClaw
CactusOnFire 21 hours ago [-]
I downloaded Hermes CLI previously. I didn't like that the default personality was "anime girl". While I like the idea of agent memory I worry it becomes less portable, and it's fundamentally more of my context window consumed.
satvikpendem 1 days ago [-]
Nous Research seems pretty cool in terms of creating open source models that are American made, which is sorely lacking. This app however, not sure why I'd use it over something like Unsloth Studio which is OSS and has more features and has the benefit of being able to use any LLM I want.
hadlock 23 hours ago [-]
Hermes offers 3 main things over something like LM Studio
1. formal memory system
2. actively develops skills and tweaks them
3. some kind of cron system for recurring tasks (check email, draft replies, daily news summary, weather, whatever)
LM Studio (or at least, last time I ran it) just downloads and runs an LLM with the ability to add a simple prompt ("you are a chat bot, be friendly and keep replies to moderate length", etc etc). Hermes is a more persistent version of claude code, or a harness for claude code/codex and is pro-vendor agnostic by design.
1. formal memory system
2. actively develops skills and tweaks them
3. some kind of cron system for recurring tasks (check email, draft replies, daily news summary, weather, whatever)
LM Studio (or at least, last time I ran it) just downloads and runs an LLM with the ability to add a simple prompt ("you are a chat bot, be friendly and keep replies to moderate length", etc etc). Hermes is a more persistent version of claude code, or a harness for claude code/codex and is pro-vendor agnostic by design.