DatattentionNo-auth model APIs for testing
A place to point a client when you are testing the client, not a provider account.
Client tests fail for boring reasons: the base URL, the header, the model field, the stream flag. A local server is the usual fixture. A remote server with no account is the same fixture when you do not want to install weights.
Use a tiny prompt and a short timeout. Assert that the status is 200 and that the body has the field your client reads: message.content for Ollama chat, choices[0].message.content for OpenAI chat, predictions for a TensorFlow predict call.
This host is that fixture. No signup, no key, no quota token. The endpoints page is the list of fixtures, one per port.
On this host
A single chat call on port 8080 is enough to see the OpenAI shape. No API key.
curl http://datattention.com:8080/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.