trajrl-bench

B待确认Benchmark

智能体环境控制开放

发布方：trajectoryRL

热度0.0±0

GitHub 主页

下载量 · 30天

—

Hugging Face

GitHub Stars

—

代码仓库

论文被引

—

Semantic Scholar

跑分模型 · 30天

—

Leaderboard results

简介

TrajRL-Bench: AI agent skills benchmark. SSH sandbox with mock services, LLM judge scoring, split-half delta evaluation. Leaderboard at trajrl.com/bench