Hermes Agent Made Crystal Clear · chapter 7: Models and Providers: Bring Your Own Brain
One agent core, many interchangeable brains
2026-10-15
Managed cloud providers (OAuth login or API key) and local runtimes (Ollama or vLLM behind a custom base_url) all attach to the same Hermes core, which selects the active model through --model and --provider. Whichever you choose, the model needs a context window of at least 64,000 tokens.
Below: the paragraph from the book that builds this idea, then the diagram itself (Figure 7.1), and a recap. About a minute of reading.
Because the agent core is constant, the choice in front of you is a two-axis one: managed versus local (who runs the hardware), and cheap versus capable (how much reasoning you are paying for). The rest of the chapter walks both axes, then shows the exact config to switch.
Recap
- The idea: Managed cloud providers (OAuth login or API key) and local runtimes (Ollama or vLLM behind a custom base_url) all attach to the same Hermes core, which selects the active model through --model and --provider.
- The picture: Figure 7.1, from chapter 7 ("Models and Providers: Bring Your Own Brain") of Hermes Agent Made Crystal Clear.
- Go deeper: the chapter builds this step by step, with recipes and sources at the end.
This diagram is one of many in Hermes Agent Made Crystal Clear.
Every chapter opens with the gist, draws the hard ideas, and ends with recipes and sources.
Get the book
