Run Hermes Agent on your own GPU

Hermes Agent is Nous Research's agent. TurboLLM wires it to a local model with one command, so it runs against whatever you've loaded — no cloud key, fully offline.

Get started (one command)

From any terminal, with TurboLLM installed:

That's it. TurboLLM wires Hermes Agent to your local daemon — writing Hermes' own config/CLI to point at TurboLLM — then launches it. If it can't parse an existing config, it refuses to overwrite it and prints the manual steps instead, so your settings are never clobbered.

Recommended models

Agents lean hard on tool-calling and long context. Browse the model catalog for good local picks by GPU size, and load one with a generous context length — agentic sessions fill the context window fast. TurboLLM's auto-tune shows a VRAM-fit verdict so you can see how much context fits before you commit.

Manual setup (no magic)

The launcher just points Hermes Agent at TurboLLM's local endpoint — there's nothing it does you can't do by hand. TurboLLM exposes an OpenAI- and Anthropic-compatible API, so configure Hermes with:

• Base URL: http://localhost:6996
• API key: your TurboLLM API key (or any non-empty value if auth is off)
• Model: the model you've loaded

Get a key from Developer → API Keys → Create — required only if you've turned on Require API key under Settings → Network (e.g. for LAN sharing).