Use DaoXE in Raycast AI

Raycast's Custom Providers reads ~/.config/raycast/ai/providers.yaml and registers whatever models you declare into every AI feature — chat, Quick AI, AI Extensions. Point one entry at DaoXE and the launcher's model picker spans GPT, Claude, Gemini, DeepSeek and your whole account catalogue on one balance (cheapest APIs guide).

Why route Raycast AI through DaoXE#

• AI in the keystroke layer. The same models everywhere Raycast asks for one — no app switching.
• One key, many vendors. Cross-vendor gateways are an explicitly supported shape, not a hack.
• Direct routing. Requests go from your device straight to the endpoint — Raycast's servers are never in the path.

Set it up in Raycast AI#

1. Open the Custom Providers section. Raycast → Settings → AI → scroll to Custom Providers → Reveal Providers Config. Copy providers.template.yaml to providers.yaml.
2. Declare the provider. Add an entry with base_url: https://daoxe.com/v1 (no /chat/completions suffix) and your key under api_keys.
3. Register exact model IDs. Each model needs an id copied verbatim from GET /v1/models, a display name, and a realistic context. Guessed IDs fail at request time.
4. Set abilities honestly. Declare what the model actually supports — tools for AI Extensions, vision for image input. Save; Raycast hot-reloads the file with no restart.

# ~/.config/raycast/ai/providers.yaml
providers:
  - id: daoxe
    name: DaoXE
    base_url: https://daoxe.com/v1
    api_keys:
      daoxe: sk-you…-key
    models:
      - id: <exact-model-id>
        name: GLM 5.3 Flash
        provider: daoxe
        context: 128000
        abilities:
          tools: {supported: true}
          vision: {supported: true}

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 client, 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."}]
  }'

Frequently asked questions#