DatattentionPointing an agent at a remote self-hosted model
Set the base URL, leave the key empty, and send one chat request.
An agent client usually has three settings: a base URL, an API key, and a model name. For an OpenAI-compatible server the base URL ends in /v1. The key can be blank when the server does not check one. The model name is the string you would send in the JSON.
On this host the base URL is http://datattention.com:8000/v1. The same client can use http://datattention.com:8080/v1 or http://datattention.com:4000/v1. For an Ollama-native client the base URL is http://datattention.com:11434 and the model is llama3.2. OpenClaw is http://datattention.com:18789/v1 with model openclaw/default. Hermes is http://datattention.com:8642/v1 with model hermes-agent.
Confirm with one chat request before you turn on tools or a long loop. If that call returns a message, the base URL is right. The endpoints page lists every route.
On this host
No API key. This is the chat request the agent settings above should produce.
curl http://datattention.com:8000/v1/chat/completions \
-H 'Content-Type: application/json' \
-d '{"model":"gpt-4o-mini","messages":[{"role":"user","content":"Hello"}]}'The full list of ports and paths is on the endpoints page.