zeph-bench

Benchmark harness for evaluating Zeph agent performance on standardized datasets.

Overview

Provides a CLI-driven benchmark runner that feeds standardized task datasets through the Zeph agent loop and records latency, token usage, and correctness metrics. Integrates with zeph-core and zeph-llm to exercise the full inference path under controlled conditions.

Installation

[dependencies]
zeph-bench = "0.20"

Or via cargo add:

cargo add zeph-bench

Note: Requires Rust 1.95 or later.

Documentation

Full documentation: https://bug-ops.github.io/zeph/

Part of the Zeph workspace.

License

MIT