Show HN: BitRun v0.2 – Run 8B ternary LLMs on CPU with live wattage telemetry

Category: ai-ml

Tags: llm-inference, cpu, quantization, cli, tui, telemetry

Score: 6.5/10 (Innovation: 6, Technical: 7, Documentation: 8, Utility: 5)

BitRun is a Go-based runner and terminal dashboard for running 1-bit/1.58-bit ternary LLMs on commodity CPUs, with live power and performance telemetry. It packages native inference engines like BitNet and Falcon3 into a zero-dependency CLI with model management and an OpenAI-compatible server. The project is interesting for its focus on energy-aware local inference and its polished Cockpit TUI.

Target audience: backend devs, ML engineers, edge computing enthusiasts

Repository: https://github.com/IshekKhal/bitrun · Go · MIT · 8 stars

View on Hacker News