DaoXECursor's model settings expose an Override OpenAI Base URL toggle plus a custom-model list. Turn it on, point the base URL at DaoXE, and add exact model IDs — Cursor's chat and inline edit will call your key. Be clear-eyed about scope: the override reroutes OpenAI-style calls, not Cursor's own Tab or server-side features. For the cost angle across both Cursor and Claude Code, see the Cursor & Claude Code guide.
Why route Cursor through DaoXE#
- Keep Cursor's UX, swap the backend. Chat and Ctrl/Cmd-K edits run on your DaoXE models instead of per-vendor accounts.
- One key, many models. Add several DaoXE IDs and switch between GPT, Claude, Gemini and DeepSeek in the model picker.
- Route by cost. A cheap model for quick edits, a frontier one for hard reasoning — one balance. See the cheapest APIs guide.
Set it up in Cursor#
- Open Model settings. Cursor Settings → Models. Scroll to the OpenAI API Key section.
- Add a custom model. Under the model list, Add model and type a name that matches an exact ID from
GET /v1/models(guessed names fail). - Override the base URL. Expand OpenAI API Key, paste your DaoXE key, enable Override OpenAI Base URL, and set
https://daoxe.com/v1. - Verify, then select it. Click Verify so Cursor sends a live request; then pick your custom model in the chat/composer model picker and run a small task.
Caveats worth knowing
The override is global to OpenAI-style calls and account-wide, not per-model: enabling it reroutes Cursor's OpenAI requests to DaoXE. Cursor's own Tab autocomplete and some server-side/agent features are tied to Cursor's infrastructure and won't use your endpoint — custom models mainly power chat and edits. Agentic edits need a tool-capable model. Use exact IDs from GET /v1/models; if Verify fails, the base URL or key is wrong. More clients in the 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}"A cheaper endpoint only helps if it serves the real model. Prove connectivity outside Cursor first, then diff a fixed 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 Cursor work with a custom OpenAI-compatible endpoint?
Yes — enable Override OpenAI Base URL in Settings → Models, set https://daoxe.com/v1, add your key and an exact model ID, then Verify.
Will Tab autocomplete use DaoXE?
No — Cursor Tab and some server-side features run on Cursor's own infrastructure. The override affects OpenAI-style chat/edit calls, so add custom models for those.
Why does Verify fail?
Almost always a wrong base URL (must end at /v1) or key, or a model name that isn't an exact ID from GET /v1/models.
Can I keep several models?
Yes — add multiple exact IDs and switch in the model picker; all run on one DaoXE balance. Current price advantage: typically ~30–80% below list via low-price groups.
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.