DaoXEEverything Continue does is declared in ~/.continue/config.yaml. Add a model with provider: openai and an apiBase pointing at DaoXE, and it routes chat/edit/apply through your chosen model. See the cheapest APIs guide for how to split cheap and frontier models by role.
Why route Continue through DaoXE#
- One file, full control. Models, roles and providers all live in
config.yaml. - Roles route work. Put a cheap DaoXE model on
autocompleteand a frontier one onchat/edit. - One key, many vendors. No per-vendor accounts for GPT, Claude, Gemini, DeepSeek or Kimi.
Set it up in Continue#
- Open your config.yaml. Edit
~/.continue/config.yaml(global) or.continue/config.yaml(per repo). - Add a DaoXE model. Use
provider: openai, setapiBase: https://daoxe.com/v1, paste your key and an exactmodelID. - Assign roles. Add
roles: [chat, edit, apply](andautocompleteon a small model) to control what each model handles. - Reload config. Click Reload config in the Continue extension, then test a chat and an edit.
# ~/.continue/config.yaml (global) — or .continue/config.yaml per repo
name: DaoXE
version: 0.0.1
schema: v1
models:
- name: DaoXE - your model
provider: openai
model: YOUR_EXACT_MODEL_ID # from GET /v1/models
apiBase: https://daoxe.com/v1
apiKey: YOUR_DAOXE_KEY
roles: [chat, edit, apply]Caveats worth knowing
Use YAML — the legacy config.json still loads but is deprecated. For gpt-5/o-series, Continue defaults to the /responses endpoint; if you hit "organization must be verified" or streaming errors, add useResponsesApi: false to force /chat/completions. The apiBase must include /v1. Roles decide which model runs each task.
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}"After reloading, prove the endpoint independently, 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#
Which provider do I pick for DaoXE?
Use provider: openai and set apiBase: https://daoxe.com/v1 — DaoXE is OpenAI-compatible.
config.yaml or config.json?
YAML. The old JSON config still loads but is deprecated; new options target YAML.
Reasoning models throw errors — why?
Continue may use /responses for gpt-5/o-series. Set useResponsesApi: false to force /chat/completions.
Can one model do autocomplete and another chat?
Yes — that's what roles are for. Put a cheap DaoXE model on autocomplete and a stronger one on chat/edit.
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.