Hermes Agent

[video] I'm building H3rm35, a native iPhone + iPad companion for Hermes Agent. Chats, multi-bot groups and multiple hosts — your agents stay on your machines. Real app UI with sample data below. Internal TestFlight for now. What would you want to control from your phone?

[image] Counting down to 100,000 stars… Hermes Agent FTW

[text] hermes agent is already the best on local models. but i'm working on more edges to make it fly even harder. before that, if your agent keeps crashing on local inference here's what to check: > max_turns: default is tuned for fast frontier models. bump from 30 to 50. slow local

[image] Gratitude to @aronprins for contributing a huge sleek UI refresh for Hermes WebUI! 🙏 And big thanks to @gabogabucho for creating a sweet end to end onboarding flow for new users right through the web UI! 🙏 If you haven’t yet there has never been a

[image] are you on v0.5.0 too?

[image] teknium just shipped 7 pluggable memory providers. this is massive. your agent can now remember you across sessions with the backend YOU choose. run 'hermes update' right now and then 'hermes memory setup' to pick your provider. if you're on local only, holographic uses SQLite

[video] Hermes Hackathon Submission: Allmind is an all in one business creation and management platform built from the ground up on Hermes. Watch me make a @stripe connected business in 16 minutes: @nousresearch @teknium

[image] BEWARE! And please share. We are all AI consumers, and this should not be acceptable for anyone 🚨

[image] i want to share something i've been working on since this recent past weekend. its a TUI for Hermes. think of it as a companion. i made this for me personally to keep track of what Hermes was learning and how it was self improving. after building it in public, people grew

[image] The `--jq=EXPR` option has proved such an excellent feature since we landed it just a week ago! Astra and other frontier models picked it up from `--help` right away without any need to mention it in skills, and are post-processing with the built-in jq all day long for all kinds

[image] Hermes Agent + self-hosted container + Claude Opus 4.6 — the vibes are immaculate

[video] The latest @ASIMOV_Platform release lands a built-in `--jev=NOUL` output filter, enabling agentic harnesses such as Hermes to offload simple natural language postprocessing decisions to Jev, a System One model Look ma, here just casually turning my LinkedIn network into JSON-LD

[text] Stop WASTING tokens. Put the scripts into play ⬇️

[text] Everything you need to know to build an Inbox Manager Agent ⬇️

[image] Test driving our social media intelligence layer in Hermes

[image] The upcoming @ASIMOV_Platform release adds Jev noul filtering to speed up complex agentic use cases Postprocessing large resource collections with both `--jev=NOUL` and `--jq=EXPR` significantly improves token efficiency in Hermes and similar harnesses

[text] For anyone running one or more Hermes profiles, the new Desktop Kanban plugin is worth a look. Create tasks, assign profiles, choose models and skills, set dependencies, and manage the workflow from one board.

[text] Hi @Teknium, can we have one button on the Hermes Agent desktop app that updates the backend, and then automatically updates the client afterwards?

[image] This past week my Hermes agent and I have been building a pure p2p messaging app utilizing Arti to embed tor directly into the app . Its in alpha stages but messaging works on both mobile and Linux. If you're interested check out the repo .

[text] Hey, based on my MAB (my a** benchmarking), I started tokenmaxxing in 4 days now instead of 2, so I achieved 2x enhancements in the cost efficiency 😎

[text] Here is my Hermes configuration for using a frontier model such as GPT-6 Astra as the main orchestrator, while cheaper or local models handle bounded subagent work. The main model owns planning, architecture, integration, and final acceptance. Qwen handles repository inspection,

[image] Today, I will add this model to my hermes config.yaml as a delegation model 💪

[text] Hi folks. Totally missed this community. Hope you all are doing well. Hermes rocks. Stay safe and see you later.

[text] Anyone is already working on optimizing this locally?

[image] Tired of babysitting Hermes during long-running tasks? I built telegram-notify: get Telegram alerts when Hermes finishes, fails, or actually needs your approval — while filtering subagent noise. Now available in the official Hermes Plugin Catalog:

[image] I built a small tool for the gap between updating Hermes and reviewing your custom skills. Hermes Skill Drift Check compares two Git revisions and points to skill lines that reference changed commands, options, tool schemas or source files. It runs locally and doesn't execute or

[text] I just made my Hermes agent to cleanup my disk, and it did a pretty well job, saved me a lot of hours figuring out where is what and what are their sizes

[text] Some numbers to share in general for better bottleneck understanding of llm hosting!

[text] Can we use our Hermes agent dynamically switching between models based on our request? I can set model globally and at a session level. What I’m asking is, I want to switch between models for the tasks I’m asking to my agent like “review this PR using Astra” then when it comes