From fbefc61098eacd93b63bb158915bc29fca824dbc Mon Sep 17 00:00:00 2001 From: seven7763 <246023385+seven7763@users.noreply.github.com> Date: Wed, 15 Jul 2026 20:34:58 +0800 Subject: [PATCH] docs: DaoXE example for OpenAI-compatible custom providers Add a short multi-model gateway example (DaoXE, https://daoxe.com/v1) under Configuring OpenAI-Compatible Endpoints. Account-scoped model IDs. Signed-off-by: seven7763 <246023385+seven7763@users.noreply.github.com> --- .../playground/02-custom-providers.mdx | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/docs/docs/prompt-engineering/playground/02-custom-providers.mdx b/docs/docs/prompt-engineering/playground/02-custom-providers.mdx index 5262335c3c..5a0134ff7c 100644 --- a/docs/docs/prompt-engineering/playground/02-custom-providers.mdx +++ b/docs/docs/prompt-engineering/playground/02-custom-providers.mdx @@ -209,6 +209,23 @@ Base URL: https://your-api-endpoint.com/v1 Make sure to include `/v1` at the end of the Base URL. ::: +### Example: multi-model OpenAI-compatible gateway (DaoXE) + +[DaoXE](https://daoxe.com) exposes an OpenAI-compatible Chat Completions API. Use the same **OpenAI-compatible** custom provider form: + +``` +API Key: +Base URL: https://daoxe.com/v1 +``` + +Then add one or more **exact** model IDs available to your DaoXE account (`GET /v1/models` or the dashboard). Do not hardcode a static public catalog — availability is account-scoped. + +Notes: + +- This path uses OpenAI Chat Completions only. DaoXE also supports other protocols (for example Anthropic Messages) for other clients. +- DaoXE is not available in mainland China. +- Contributor disclosure: this example was contributed by a DaoXE affiliate. + :::warning If you're running both Ollama and self-hosting Agenta on your local machine, set the Base URL to http://host.docker.internal:11434/v1 to allow Agenta's completion service (running behind Docker) to access Ollama. :::