Hermes Agent Adds Asynchronous Subagents for Non-Blocking Delegation
Many users praised Hermes Agent's new asynchronous subagents for enabling non-blocking delegation, calling the update awesome and thanking the team for rapid feature shipping that fixed a prior limitation.
Most Activity
@Teknium Has anyone made a video explaining this in a visual way? I need this but I don’t understand the setup of the subagents.
@Teknium Bir de Intel Iris mac' de 'hermes-desktop' localde kurmaya izin vermiyor, m apple silicon gerekli hatası alıyorum, oraya da bir güncelleme lütfen..
No more waiting for subagents to finish to keep working with your Hermes Agent.
Asyncronous subagents are here.
Hermes Agent now supports asyncronous subagents!
The existing delegate tool, which your agent uses to spawn subagents to fan out and do work, no longer blocks your chat!
To access now, `hermes update`, and enjoy!
@Teknium https://github.com/blueberryvertigo/traitee
had this a few months ago. great feature!
@NousResearch massive w
@NousResearch This unlocks a powerful new workflow for Athena
With the latest update, the Hermes orchestrator gets a serious capability boost making context flow faster, smoother, and smarter across the system
https://github.com/luckeyfaraday/Athena
@NousResearch Every day I look in that bottom right corner and it says +106 or more. You guys are amazing
@Teknium you guys are just adding things i made in slermes now
just saying lol
@Teknium What is the prompt for the images you have for each release?
@Teknium Fucking madlad, I just saw the person mentioning it a little bit ago
@Teknium Assume i allow 4 concurrent session on my llm server, shouldnt there be a configuration possible that hermes can use 3 subagents and keep 1 to always respond to user requests?
@bowecho777 @Teknium Done.
@Teknium Does this mean other agents with profiles work parallel? Or does it mean it spawns a temporary sub agent to work parallel? Just asking for my app, it detects agents, but not background temporary agents
@Teknium ASYNC BACK (!)
@Teknium Hit this all the time with bots and robotics. Query multiple sources, slowest one blocks everything.
Does this handle timeout guarantees? Like, can you say "wait max 500ms, then aggregate what you got"? That's what kills production at scale.
@Teknium Please implement Fusion API by open router plsplspls
@Teknium Nice update. Thanks for all the work on Hermes. I’ve got bunches of them running for myself, my wife’s company. Even my kids have them!
I’m looking forward to when Hermes desktop can connect to hosted Hermes and control a local computer.
@waefrebeorn @Teknium slermes ?
@Teknium Ok this is BIG
@Teknium oh word