EvalCore
The deterministic eval engine for AI systems
什麼是 EvalCore？
Measure, compare, and regression-gate LLM apps and agents — one config-first binary. Statistical trials, side-by-side model comparison, agent trajectories from OTel traces, a local run viewer, and record/replay caching for deterministic $0 CI. Early-stage & open source, feedback shapes the roadmap.
截圖
?
還沒有評論，來搶沙發吧！
X 上關於 EvalCore 的真實討論
去 X 發文