Agent infrastructure
robodojo
robodojo-benchmark/robodojo
A unified sim-and-real benchmark for generalist robot manipulation with tasks, simulation, and leaderboards.
Topics
Related projects
opencompass7.6kA one-stop open platform for evaluating foundation models across 100+ datasets, with unified inference, judging, and leaderboards.
OSWorld3.3kOSWorld is a benchmark and real desktop environment for evaluating multimodal computer-use agents, with support for VMware, Docker, AWS, and other backends for reproducible task evaluation.
adr2.1kAn enterprise security system that observes, benchmarks, and defends AI agents in production.
cl-bench592A benchmark series for context learning that evaluates reasoning and learning ability through workplace and everyday tasks with thousands of detailed annotation rules.
hyperfine29.3kA command-line benchmarking tool that runs commands repeatedly and reports statistical timings and hardware metrics, with warmups, parameter scans, and exports.
text-to-cad26.1kA plugin and skill pack that gives coding agents CAD workflows, letting them generate STEP, STL, GLB, and 3MF models and run design-for-manufacturing checks.