Qubrid AI for Dify

OpenAI-compatible serverless models from the Qubrid AI Platform.

Source repository: github.com/QubridAI-Inc/Dify-Plugin

Contact: [email protected]

Setup

1. In Dify, go to Plugins → Install plugin → Install from local package.
2. Upload (or install from the Dify Marketplace once published).
3. Go to Settings → Model Provider → Qubrid AI → Set up.
4. Enter your API key from platform.qubrid.com/api-keys.
5. Leave API Base URL empty to use the default (), or set it explicitly.
6. Save, then select a Qubrid model in any Dify app or workflow.

Usage

The plugin registers Qubrid AI as a model provider. Use it anywhere Dify supports LLMs: chat apps, agents, and workflows.

Predefined models are listed in the model picker. You can also add any other Qubrid model ID as a custom model.

Required credentials

Connection requirements

Your Dify instance must be able to reach:

• (model inference and catalog API)

Outbound HTTPS (port 443) is required. No inbound ports or webhooks are needed.

Included models

The plugin ships 43 predefined chat models synced from the Qubrid catalog (), including:

You can also add any other Qubrid model ID as a custom model.

Privacy

See PRIVACY.md [blocked] for what data this plugin sends to Qubrid and how credentials are handled.

---

Developer guide

These steps are for maintaining the plugin source, not for end-user installation.

Project layout

Refresh models from catalog

E2E tests (local Dify)

Requires local Dify at and API keys configured for testing.

Rebuild the package