Artificial intelligence helps streamline modeling, coding, and architectural workflows, but many organizations require specific custom large language model (LLM) providers or strict API configurations to align with security, performance, or cost policies. Visual Paradigm supports custom LLM endpoint integration, enabling teams to connect their own API keys and language models to power AI features across both cloud and desktop tools.
Key Benefits of Connecting Your Own LLM Endpoint
Connecting a custom LLM provider offers several advantages for system architects, developers, and enterprise teams:
- Full Provider Choice: Bring your own preferred AI model providers and leverage custom API keys directly within your workspace.
- Cross-Platform Synchronization: Configure settings once in the web workspace, and connected desktop installations instantly utilize the same LLM setup.
- Advanced Parameter Control: Adjust fine-tuned AI behaviors, such as thinking budgets and reasoning levels, for models that support parameter customization.
- Verified Quality Assurance: Easily identify models tested and verified by Visual Paradigm to ensure optimal performance and output formatting.
Step-by-Step Guide: Configuring Custom LLM Settings
Step 1: Access the LLM Configuration Panel
Log in to your workspace in Visual Paradigm Online. Launch an AI-driven tool, such as the AI Chatbot Tool or VPasCode Editor. Click your avatar icon in the top right corner of the interface and select LLM Configuration from the menu.
Step 2: Select Provider and Input API Key
In the configuration dialog, choose your preferred LLM provider from the list. Paste your valid API key into the designated field to establish authentication.
Step 3: Choose and Verify Your LLM Model
Select the specific model you wish to use from the model selection dropdown. If you select a model that has not yet been verified, a notification warning will appear. Expand the Verified Models section to browse models tested and approved for consistent output quality.
Step 4: Configure Advanced Model Behaviors (Optional)
If your selected model supports additional inference parameters, configure options such as Thinking Budget or Thinking Level to control reasoning depth for complex tasks.
Step 5: Save and Apply Settings Across Tools
Click Save Changes to apply your configuration. Your selected LLM will immediately begin powering AI tools, including Chatbot utilities and VPasCode. Furthermore, if you use the desktop application connected to the same online repository, local AI diagram generation features will automatically route requests through your configured custom LLM endpoint.
Supported Applications and Capabilities
- Cloud AI Tools: Enhances interactive assistant features in web applications and code generators.
- Desktop Diagram Generation: Bridges cloud-defined LLM settings to local client software for AI-assisted UML and system modeling.
- Documentation Platforms: Integrates with writing workflows inside the OpenDocs environment for automated report generation.
Take Control of Your AI Workflows
Unify your AI configuration across cloud and desktop environments for greater flexibility and control over system modeling.
Get started with Visual Paradigm tools today:
