U9 AI · DOCUMENTATION
Dify setup
Configure Dify with the correct U9 address, model and compatibility checks.
Download Markdown ↓Compatibility and prerequisites#
This guide prepares Dify integration with U9 AI. You need an active U9 key, sufficient credit and an available model matching the endpoint. Official documentation was checked on 2026-10-24; this is not an end-to-end test of every client version or upstream model.
Setup steps#
- Have a workspace administrator install or enable the OpenAI-API-compatible model provider plugin.
- Add an LLM using the exact U9 model ID; the display name can be U9 AI.
- Enter the U9 /v1 API Base URL and your U9 API Key.
- Set real context/output limits and tool capabilities. Save and validate, then select the model in an application or workflow LLM node.
Configuration values#
| Setting | Value |
|---|---|
| Base URL | https://u9ai.ai/v1 |
| API Key | YOUR_U9AI_API_KEY · U9 API key |
| Model | YOUR_CHAT_MODEL_ID · Available model ID |
| Expected chat endpoint | https://u9ai.ai/v1/chat/completions |
Important differences#
This configures a text LLM. Knowledge-base embedding and rerank require separately available models; asynchronous video is not an LLM node. Shared workspace usage belongs to the configured U9 key.
Acceptance and handoff#
First follow quickstart to check GET /v1/models without generating content. Then test a short non-streaming request, streaming and any required tool-call round trip separately. Generation tests may incur charges. Record the request ID and verify cost analysis. For failures, include client version, model ID, endpoint, status and sanitized errors in a ticket. Do not blindly resubmit timed-out jobs.
Give this Markdown file to your developer or agent. Ask it to preserve current settings, substitute only available model IDs and report untested capabilities. Never put real keys in shared Markdown.