Ollama (Local)
Ollama Local Integration (April 2026)
Section titled “Ollama Local Integration (April 2026)”AgentV supports local model evaluation via Ollama, enabling private, high-speed benchmarking for open-source models like Llama 4.
🚀 Native Protocol
Section titled “🚀 Native Protocol”Use the ollama protocol to connect to a local Ollama instance.
agentv run --path scenarios/loan_scenario.json --protocol ollama --agent ollama://llama4Configuration
Section titled “Configuration”By default, the harness attempts to connect to http://localhost:11434. You can override this in your .env:
OLLAMA_HOST=http://your-server:11434OLLAMA_MODEL=llama4🛠 Supported Models
Section titled “🛠 Supported Models”AgentV is optimized for the 2026 local model landscape:
llama4(8B, 70B, 400B)mistral-v4phi-5
🧪 Benchmarking Local vs. Cloud
Section titled “🧪 Benchmarking Local vs. Cloud”A common research use case is comparing a local model against a cloud baseline (e.g., Llama 4 vs. GPT-5.4).
Use the Model Wars protocol:
agentv run --path scenarios/loan_scenario.json --compare agents_inventory.yaml⚠️ Performance Note
Section titled “⚠️ Performance Note”Local execution speed depends on your hardware (GPU/RAM). If you experience timeouts during evaluation, increase the DEFAULT_ADAPTER_TIMEOUT in your config.