Use DaoXE in NextChat

NextChat speaks OpenAI-compatible HTTP, so DaoXE is an endpoint swap — in the Settings UI or via OPENAI_BASE_URL when you self-host. Its CUSTOM_MODELS list then shapes exactly which account-scoped IDs appear (cheapest APIs guide).

Why route NextChat through DaoXE#

• Self-hosted in minutes. Docker + two env vars; your conversations live on your server.
• The picker is your catalogue. CUSTOM_MODELS whitelists (or --prefixes away) real IDs from GET /v1/models.
• One key for the whole instance. Every user of your NextChat deployment shares the DaoXE balance.

Set it up in NextChat#

1. Hosted app: Settings. Open Settings → Model Provider → OpenAI, paste the DaoXE key, and set the API endpoint to https://daoxe.com/v1.
2. Self-hosted: env. Set OPENAI_API_KEY and OPENAI_BASE_URL=https://daoxe.com/v1 in docker-compose / Vercel env.
3. Shape the model list. Use CUSTOM_MODELS to list exact IDs (prefix - to hide built-ins you can't call).
4. Send a test message. Then check the response's model field — it should echo the ID you picked.

# self-hosted: docker-compose env
OPENAI_API_KEY=sk-you…-key
OPENAI_BASE_URL=https://daoxe.com/v1
CUSTOM_MODELS=glm-5.3-flash,-gpt-4o

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}"

Verify the endpoint before wiring the UI, 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#