Use DaoXE in Kilo Code

Kilo Code speaks to any OpenAI-compatible endpoint out of the box, so DaoXE is a provider setting — base URL and key. Like Roo, it lets different modes (Architect, Code, Ask) run different models, which is where a single multi-vendor key earns its keep: cheapest APIs guide.

Why route Kilo Code through DaoXE#

• Per-mode models. Architect on a frontier model, Code on a workhorse — one key, one balance.
• Cline/Roo lineage. If you've configured either, the settings layout will feel familiar.
• Verified by docs. Kilo's own docs treat custom OpenAI-compatible base URLs as a first-class path — local servers, corporate proxies, gateways.

Set it up in Kilo Code#

1. Open provider settings. In the Kilo Code panel, open settings and set API Provider to OpenAI Compatible.
2. Set the Base URL. https://daoxe.com/v1 — the /v1 suffix must be there (Kilo's troubleshooting doc lists a missing /v1 as the classic 404 cause).
3. Paste key and Model ID. Your DaoXE key, plus a Model ID copied verbatim from GET /v1/models.
4. Assign models per mode. Optional but the point: give Architect, Code and Ask modes different models to trade cost for quality where it matters.

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}"

Connectivity first, then prove the model is real — same hard prompt, temperature 0, diff against the official API:

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."}]
  }'

Frequently asked questions#