benchmarking

Local-first experiment and evaluation workbench plugin for DeepSeek Harness (DSH).

DSH plugin: run a command N rounds, judge by median/distribution — 批量回归取统计结论

Reproducible paired A/B experiments for DeepSeek Harness plugins, with isolated runners, auditable evidence, and deterministic promotion gates.

Install with npm i dsh-benchup. Reproducible, profile-aware benchmarks for DeepSeek Harness — compare models, plugins, prompts, and agent strategies.