DaoXEBoltAI ships built-in providers (OpenAI, Anthropic, OpenRouter…) but its generic escape hatch is the OpenAI-compatible Server type under Settings → Models. DaoXE speaks that shape, so one entry covers the whole catalogue (cheapest APIs guide).
Why route BoltAI through DaoXE#
- Native macOS everywhere. Chat, AI Command and AI Inline all ride the same provider — one DaoXE key powers shortcuts in any app.
- One key, many models. Switch models from the chat-configuration popover instead of juggling vendor accounts.
- Local + remote side by side. BoltAI's Ollama support and a DaoXE provider entry share one model picker.
Set it up in BoltAI#
- Open the provider form. Go to Settings → Models, click the (+) button and choose OpenAI-compatible Server.
- Name and endpoint. Give it a friendly name, then paste the exact chat completions URL:
https://daoxe.com/v1/chat/completions. BoltAI's own guide says to "enter the exact url for the chat completions endpoint" — its LM Studio example ishttp://localhost:1234/v1/chat/completions. - Model ID and context. Enter the model ID that gets sent as the
modelparameter (one default per entry — copy an exact ID fromGET /v1/models), plus its context length from the model's docs. - Streaming, default, save. Enable streaming if you want it; click Save Changes. BoltAI's docs flag the last step: "if you don't intend to use OpenAI, make sure to set this as default."
Caveats worth knowing
The endpoint is a full path: unlike most clients, BoltAI's OpenAI-compatible form wants the complete /chat/completions URL, not a base URL — a bare /v1 will not work here. One model per entry: the form takes a single model ID (it "will be sent with each chat request"); add another entry per model you need. Context length is on you: the docs say to "refer to the original model" — wrong values waste budget or truncate. Default service matters: AI Command with a custom server is beta and only fires when that server is the default AI service. Keychain prompts: keys are stored in the Apple Keychain — if they stop persisting, re-grant BoltAI access in Keychain Access (entry BoltAI_openAIKey). 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}"Prove the endpoint outside the app, 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#
Can BoltAI use DaoXE?
Yes — Settings → Models → (+) → OpenAI-compatible Server with https://daoxe.com/v1/chat/completions and your DaoXE key.
Base URL or full endpoint?
Full endpoint — BoltAI asks for the exact chat completions URL (…/v1/chat/completions), which differs from most OpenAI-compatible clients that take a base URL.
Why does my custom server not fire in AI Command?
BoltAI documents AI Command with a custom server as beta: set the OpenAI-compatible entry as the default AI service, or commands keep using the old provider.
What does it cost?
Per-model, account-scoped — 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.