Developer Documentation
EverestQ Engine Quickstart
EverestQ Engine is a production-grade distributed AI runtime designed for zero-latency LLM inference, local model management, and cloud cluster deployment.
PRODUCTION TIP
EverestQ Engine listens on
http://localhost:11434 by default, offering 100% OpenAI REST API endpoint compatibility out of the box.CLI Engine Commands
Terminal Commandsbash
# List all locally loaded models and VRAM usage eq models # Execute inference with custom temperature and system prompt eq run everestq-coder "Write a concurrent worker pool" --temperature 0.7 # Inspect distributed cluster topology eq nodes # Perform systems research experiment with locked seed ($seed=42$) eq research run --name "GQA Attention Benchmark"
HARDWARE INSPECTION
On cold daemon start, EverestQ automatically detects CUDA GPUs and Apple Silicon Metal accelerators to optimize memory page block allocations.