From 3accaf0a5b7f617204830d33b2b015e89c937df3 Mon Sep 17 00:00:00 2001 From: seven7763 <246023385+seven7763@users.noreply.github.com> Date: Wed, 15 Jul 2026 01:49:53 +0800 Subject: [PATCH 1/2] docs: add DaoXE OpenAI-compatible Chat Completions backend example Co-Authored-By: Claude Fable 5 --- massgen/configs/BACKEND_CONFIGURATION.md | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) diff --git a/massgen/configs/BACKEND_CONFIGURATION.md b/massgen/configs/BACKEND_CONFIGURATION.md index 0a31f0ad1..4d9422b61 100644 --- a/massgen/configs/BACKEND_CONFIGURATION.md +++ b/massgen/configs/BACKEND_CONFIGURATION.md @@ -328,7 +328,7 @@ backend: Generic backend supporting multiple providers (v0.0.18+ with MCP). -Supports: Cerebras AI, Together AI, Fireworks AI, Groq, Nebius AI Studio, OpenRouter, Kimi/Moonshot, and any OpenAI-compatible API. +Supports: Cerebras AI, Together AI, Fireworks AI, Groq, Nebius AI Studio, OpenRouter, Kimi/Moonshot, DaoXE, and any OpenAI-compatible API. ```yaml backend: @@ -365,6 +365,20 @@ backend: - **OpenRouter**: `https://openrouter.ai/api/v1` - **Nvidia NIM**: `https://integrate.api.nvidia.com/v1` - **Kimi/Moonshot**: `https://api.moonshot.cn/v1` +- **DaoXE**: `https://daoxe.com/v1` (multi-model multi-protocol gateway; Chat Completions path; account-scoped model IDs; not available in mainland China) + +### Example: DaoXE multi-protocol gateway + +[DaoXE](https://daoxe.com) is a multi-model multi-protocol API gateway. Use the Chat Completions backend with account-scoped model IDs from your DaoXE dashboard (`GET /v1/models`). Not available in mainland China. + +```yaml +backend: + type: "chatcompletion" + model: "YOUR_DAOXE_MODEL_ID" + base_url: "https://daoxe.com/v1" + api_key: "${DAOXE_API_KEY}" # or omit if DAOXE_API_KEY / OPENAI_API_KEY is set +``` + --- From 640ab0ebddd17fdf0147041a43bd2e8c7b3ad7a1 Mon Sep 17 00:00:00 2001 From: Seven <88984809@qq.com> Date: Thu, 16 Jul 2026 05:05:58 +0800 Subject: [PATCH 2/2] docs: clarify DaoXE example uses OPENAI_API_KEY fallback --- massgen/configs/BACKEND_CONFIGURATION.md | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/massgen/configs/BACKEND_CONFIGURATION.md b/massgen/configs/BACKEND_CONFIGURATION.md index 4d9422b61..834f1add0 100644 --- a/massgen/configs/BACKEND_CONFIGURATION.md +++ b/massgen/configs/BACKEND_CONFIGURATION.md @@ -376,7 +376,17 @@ backend: type: "chatcompletion" model: "YOUR_DAOXE_MODEL_ID" base_url: "https://daoxe.com/v1" - api_key: "${DAOXE_API_KEY}" # or omit if DAOXE_API_KEY / OPENAI_API_KEY is set + # Chat Completions backend falls back to OPENAI_API_KEY when api_key is omitted. + # Prefer an explicit key if you keep multiple OpenAI-compatible backends configured: + api_key: "${OPENAI_API_KEY}" +``` + +Export a DaoXE key into the env var the backend actually reads: + +```bash +export OPENAI_API_KEY="" +# optional if your process does not load the YAML value above: +# export OPENAI_BASE_URL="https://daoxe.com/v1" ```