Ollama Integration

Setup Ollama

1. Install Ollama: Follow the instructions on the official Ollama website to install the required software.
2. Run the Model: Ensure the desired model is available and running.

Example Command to Serve the Model

To serve the VLLM model “microsoft/phi-4”, use the following command:

ollama pull "phi4"
ollama serve "phi4"

Run Noir with Ollama

To leverage Ollama capabilities for additional analysis, use the following command:

noir -b ./spec/functional_test/fixtures/hahwul \
     --ai-provider=ollama \
     --ai-model=phi4

This command performs the standard Noir operations while utilizing the specified Ollama model for enhanced analysis.