1 Run Ollama to create the variant
~/.local/bin/hermes model
link:
https://hermes-agent.nousresearch.com/docs/integrations/providers
2 Choose
Told chat options says scroll down and choose Custom endpoint (enter URL manually)
3 local ollama url address
This is where we enter the Ollama local address,
http://localhost:11434
from previous lesson and we add /v1 at end
http://localhost:11434/v1
ENTER NOTHING for API key
4 Choose
Choose 2. Chat Completions Use /chat/completions for ...
then 1. for
qwen3.5-4b-64k:latest
Or whichever is your variant
5 Choose
enter,
64000
for context length,
press "Enter" To keep [localhost]
6 Choose
medium for reasoning effort for now
Run Hermes Desktop with
~/.local/bin/hermes desktop
link to docs,
https://hermes-agent.nousresearch.com/docs/user-guide/desktop
5 Starting Hermes desktop
run in One Terminal
ollama server
And in a second terminal run,
~/.local/bin/hermes desktop
Everything is working, we don't need to adjust
press "Enter" To keep [localhost]