SETUP GUIDE

OpenAI-compatible (generic)

Any tool that speaks the OpenAI API can connect to Profundo. Point it at the base URL and it works.

Status: Supported

1. Recommended setup

After signup, verification, and membership activation, issue a key in the dashboard. It is displayed once; keep it in a local secret store.

# Any OpenAI-compatible client can connect to Profundo.
# Just point it at the base URL and use your API key.

base_url = "https://api.profundoai.com/v1"
api_key = "pf_..."   # issue one in your dashboard

# Current model IDs are listed by GET /v1/models.

1. Create an account and verify your email.
2. Issue an API key in the dashboard (shown once; store it safely).
3. Point your OpenAI-compatible client at https://api.profundoai.com/v1 and use your key as the bearer token.
4. Pick a model from the catalog list above (also available from GET /v1/models).

Compatibility note: Some clients add their own endpoints (embeddings, images, assistants, fine-tuning). Profundo supports text Chat Completions and the Responses API; those extra endpoints are not available.

curl https://api.profundoai.com/v1/chat/completions \
  -H "Authorization: Bearer $PROFUNDO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-6-astra","messages":[{"role":"user","content":"Say hello"}]}'

client = OpenAI(
    base_url="https://api.profundoai.com/v1",
    api_key=os.environ["PROFUNDO_API_KEY"],
)

2. Verify your key

Set PROFUNDO_API_KEY in your shell, then run:

curl https://api.profundoai.com/v1/chat/completions \
  -H "Authorization: Bearer $PROFUNDO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-6-astra","messages":[{"role":"user","content":"Say hello"}]}'

Use any model returned by GET /v1/models with https://api.profundoai.com/v1.