DaoXEEvery n8n AI node that speaks OpenAI-compatible HTTP will take DaoXE unchanged: create one credential, point its Base URL at DaoXE, and every workflow on the instance inherits the catalogue. That matters because automation is a multi-model workload by nature — triage on cheap models, generation on good ones, and the cost split in cheapest APIs guide applies node by node.
Why route n8n through DaoXE#
- One credential, whole instance. Agents, chains and routers share a single key and balance.
- Right-size each node. A classification step doesn't need a frontier model; an agent's final pass does.
- Self-hosted stays yours. Your workflows keep running on your n8n; only the inference endpoint moves.
Set it up in n8n#
- Create an OpenAI credential. In n8n: Credentials → New → OpenAI. Paste your DaoXE key and set Base URL to
https://daoxe.com/v1. - Attach it to the node. In an AI Agent / OpenAI Chat Model node, select this credential instead of the default OpenAI one.
- Set the model ID. Type an exact model ID from
GET /v1/models— the dropdown won't guess for you on a custom base URL. - Split tiers per step. Cheap model for routers and classifiers, strong model for generation and the agent loop.
Caveats worth knowing
Include /v1 in the Base URL — n8n appends the rest. Embeddings nodes need an embedding-capable model ID from your account list; a chat-only ID fails there. Agent nodes depend on tool calling, so pick a model that does it properly. If the instance is self-hosted, outbound HTTPS to the endpoint must be allowed. More clients: client setup notes.
Verify you actually get the model#
Prove the endpoint works before blaming the client — if this fails, no setting will fix it:
export DAOXE_API_KEY="your_api_key"
# List the exact model IDs your account can call
curl --fail-with-body --show-error --silent \
https://daoxe.com/v1/models \
-H "Authorization: Bearer ${DAOXE_API_KEY}"Test the credential outside any workflow first, then diff a hard prompt against the official API at temperature 0:
curl --fail-with-body --show-error --silent \
https://daoxe.com/v1/chat/completions \
-H "Authorization: Bearer ${DAOXE_API_KEY}" \
-H "Content-Type: application/json" \
-d '{
"model": "YOUR_EXACT_MODEL_ID",
"max_tokens": 64,
"messages": [{"role": "user", "content": "Say hello in one sentence."}]
}'Verify us — don't trust us
Point the open benchmark at DaoXE and at the official API and compare at temperature 0. Then learn to detect model swapping so a cheaper endpoint can't quietly swap you to a smaller model.
Frequently asked questions#
Does n8n support custom OpenAI-compatible endpoints?
Yes — the OpenAI credential has a Base URL field; set it to https://daoxe.com/v1 and every OpenAI-shaped node inherits it.
Do embeddings work too?
With an embedding-capable model ID from GET /v1/models. Chat model IDs won't work in embedding nodes.
Can each node use a different model?
Yes — that's the point. Nodes set their own model ID, so routers can run cheap and generators run strong.
What does it cost?
Per-model, account-scoped — see live pricing; top-up is a flat 1 RMB = $1 of credit on every payment method.
Try DaoXE — and benchmark it yourself
One key for GPT, Claude, Gemini, DeepSeek and more. Point the open benchmark at us and compare — don't take our word for it.