Skip to content

docs(admins): add DaoXE OpenAI-compatible LLM gateway guide - #415

Open
seven7763 wants to merge 1 commit into
onyx-dot-app:mainfrom
seven7763:docs/daoxe-openai-compatible-provider
Open

docs(admins): add DaoXE OpenAI-compatible LLM gateway guide#415
seven7763 wants to merge 1 commit into
onyx-dot-app:mainfrom
seven7763:docs/daoxe-openai-compatible-provider

Conversation

@seven7763

Copy link
Copy Markdown

Description

Adds an admin guide for configuring DaoXE as an OpenAI-compatible multi-model gateway in Onyx.

DaoXE is not a first-party Onyx provider card. The guide walks through the existing OpenAI-Compatible path under Self-hosted & Custom:

  • API base: https://daoxe.com/v1
  • Bearer API key from the DaoXE account
  • Fetch Available Models (account-scoped model IDs — no hardcoded public catalog)
  • Multi-protocol note: Onyx uses Chat Completions here; DaoXE also exposes OpenAI Responses and Anthropic Messages for other clients
  • Mainland China unavailability + use site-documented regional endpoint if needed

Also links DaoXE from the Language Models overview and the generic custom-provider page, and registers the page in docs.json.

Notes

  • Docs-only; no product/code changes in onyx-dot-app/onyx
  • I maintain DaoXE. OpenAI SDK examples: https://github.com/seven7763/DaoXE-AI
  • Pattern matches existing gateway pages (OpenRouter / Bifrost / LiteLLM Proxy) while correctly using the OpenAI-Compatible UI card

Checklist

  • Page under admins/ai_models/
  • Nav entry in docs.json
  • Overview card + cross-link from custom inference provider
  • No secrets or hardcoded model ID / price tables

Document configuring DaoXE via Onyx's OpenAI-Compatible provider card
(API base https://daoxe.com/v1, account-scoped model IDs, multi-protocol
context). Wire the page into Language Models navigation and overview.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant