Skip to main content

​
Bring Your Own LLM

This tutorial demonstrates how Corvic AI’s Bring Your Own LLM feature lets you build production-ready GenAI applications without being locked into a single model provider. Learn how easy it is to plug in your own LLM endpoints—whether they’re cloud-hosted, self-hosted, or domain-specific—while keeping full control over privacy, cost, and performance.

​
What You’ll Learn

  • Integrating custom LLM endpoints into Corvic AI
  • Using cloud-hosted, self-hosted, or domain-specific models
  • Maintaining control over privacy, cost, and performance
  • Building production-ready GenAI applications with flexibility

​
Key Concepts

​
Bring Your Own LLM

Corvic AI seamlessly integrates custom LLMs into your data pipelines and agents, enabling enterprises to use proprietary or fine-tuned models alongside open or commercial LLMs. This approach gives you maximum flexibility while ensuring your data stays secure and your architecture stays future-proof.

​
Supported LLM Types

You can integrate cloud-hosted LLMs (from various providers), self-hosted models (running on your infrastructure), or domain-specific models (fine-tuned for your use case). All while maintaining full control over privacy, cost, and performance.

​
Setting Up Custom LLMs

​
Step 1: Configure LLM Endpoint

Set up your custom LLM endpoint by providing the API endpoint URL, authentication credentials, and any required configuration parameters.

​
Step 2: Integrate with Data Pipelines

Integrate your custom LLM seamlessly into Corvic AI’s data pipelines. The platform handles the integration automatically, allowing your models to work with your embedding spaces and agents.

​
Step 3: Configure Agents

Configure your agents to use your custom LLM. You can use proprietary or fine-tuned models alongside open or commercial LLMs, giving you maximum flexibility in your agent architecture.

​
Benefits

Build production-ready GenAI applications without vendor lock-in. Maintain full control over privacy by keeping data within your infrastructure, optimize costs by choosing the most cost-effective models, and ensure future-proof architecture by easily switching or combining different LLM providers.

​
Best Practices

Choose LLM endpoints that align with your privacy, cost, and performance requirements. Test custom LLMs thoroughly before deploying to production, and consider using a combination of models for different use cases to optimize both cost and performance.

Agents

Configure agents to use your custom LLM.

Admin Console

Configure custom LLM integrations in the admin console.

Data Apps

Deploy applications using your custom LLMs.

API Integrations

Learn about integrating with various LLM providers.

​
Additional Resources