OpenViking
Long-term memory for Hermes. Capture conversations, recall useful facts across sessions and search your knowledge. Choose Quick Local, OpenViking Service (managed cloud) or your own server.
Good to know before installing
Conversation turns, tool results and MEMORY.md / USER.md edits are sent to the configured OpenViking server. Quick Local setup installs the newest OpenViking 0.5.x server release from PyPI at setup time (checked against PyPI's hash, not the version reviewed here; re-running setup upgrades it) plus a hash-pinned llama-cpp-python wheel and a local embedding model. It copies the active Hermes static LLM API key into a private file under the profile for extraction, and its localhost server keeps running after Hermes exits and after plugin removal (stop it with `hermes openviking local stop`). Extraction can use a remote LLM. Custom localhost connections can start openviking-server from PATH with the profile environment minus Hermes bot, gateway, GitHub and dashboard secrets; model and tool API keys still pass.
From the plugin's catalog listing.
How to install
- Install it. Hermes asks for any API keys it needs, then asks whether to turn it on.
hermes plugins install openviking - Said no? Turn it on later:
hermes plugins enable openviking - Restart Hermes so it loads the plugin.
Some plugins also need a sign-in or settings: see Full setup & settings. Memory plugins ask whether to become your memory provider instead. Steps follow the official plugins guide (opens in a new tab).
- Category
- Memory & notes
- Maintained by
- volcengine (Community)
- Needs
- No API keys declared
- Needs Hermes
- 0.21.4 or newer
- Version
- 2.1.2
- In the catalog
- 2026-10-06, updated 2026-10-24
- GitHub stars
- 40,927
Full setup & settings ↗Source code ↗
More in memory & notes
New Hermes Agent updates every day on X. Follow @HermesWatcher