首页 /基准、评测与校准 /jev-playground by hegargarcia
jev-playground by hegargarcia
在状态明确、合法动作明确、结果可衡量的游戏中把 Jev 与其他模型作对比。
本页由英文清单自动生成。
来自 readme
Jev Playground A playground for benchmarking TypeSafe AI’s Jev against other evaluation models in games with explicit states, legal actions, and measurable outcomes. The core question: how well does each model choose the next action when the rules and available choices are clearly defined? Games provide a small, inspectable environment for exploring …
详情
- 分区
- 基准、评测与校准
- owner
- hegargarcia
- 星标
- 0
- 复刻
- 0
- 最近提交
- 2026-09-17
- 语言
- TypeScript
这是什么
- 形态
- 数据集或基准
- 宿主智能体
- 独立运行
- 面向人群
- 研究人员
成熟度
文档