https://github.com/5queezer/hermes-agent-arm64
ARM64 build of NousResearch/hermes-agent for Coolify deployment
https://github.com/5queezer/hermes-agent-arm64
Last synced: 15 days ago
JSON representation
ARM64 build of NousResearch/hermes-agent for Coolify deployment
- Host: GitHub
- URL: https://github.com/5queezer/hermes-agent-arm64
- Owner: 5queezer
- Created: 2026-04-08T22:36:35.000Z (2 months ago)
- Default Branch: master
- Last Pushed: 2026-05-28T07:25:03.000Z (15 days ago)
- Last Synced: 2026-05-28T09:15:14.408Z (15 days ago)
- Language: Dockerfile
- Size: 14.6 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Hermes Agent ARM64
ARM64 Docker build of [NousResearch/hermes-agent](https://github.com/NousResearch/hermes-agent) for deployment on ARM64 servers (e.g. Hetzner CAX, Raspberry Pi, AWS Graviton).
The upstream image on Docker Hub is amd64-only. This repo builds a native ARM64 image via GitHub Actions.
## Image
```
ghcr.io/5queezer/hermes-agent-arm64:latest
```
## Auto-update
A GitHub Actions workflow runs daily at 06:00 UTC:
1. Checks upstream `NousResearch/hermes-agent` HEAD for changes
2. Builds ARM64 image on a native `ubuntu-24.04-arm` runner (~6 min)
3. Pushes to GHCR
4. Triggers Coolify redeploy via API
Manual trigger: `gh workflow run build.yml`
## Included extras
Only the extras needed for a Telegram gateway deployment:
`messaging`, `cron`, `cli`, `pty`, `mcp`, `voice`
The full `[all]` extras hit pip's `resolution-too-deep` limit on ARM64. Install additional extras manually inside the container if needed.
## Configuration
The image starts in **gateway mode** (`CMD ["gateway"]`). Set these environment variables:
| Variable | Required | Description |
|---|---|---|
| `TELEGRAM_BOT_TOKEN` | yes | From [@BotFather](https://t.me/BotFather) |
| `TELEGRAM_ALLOWED_USERS` | yes | Comma-separated numeric Telegram user IDs |
| `HERMES_HOME` | no | Data directory (default: `/opt/data`) |
| `HERMES_INFERENCE_PROVIDER` | no | LLM provider (e.g. `openai-codex`) |
For OpenAI Codex, run the OAuth login interactively after first deploy:
```bash
docker exec -it hermes model --no-browser
```
## Persistent data
Mount a volume at `/opt/data` for config, sessions, skills, and credentials.