Harbor Hub

HARBOR HUB

Environment-driven
agent development

Share tasks, benchmarks, leaderboards, and trajectories.
Scale to billions of rollouts.

uv tool install harbor

Explore datasets on the Hub

View all
terminal-bench
A benchmark for measuring agents’ abilities to complete tasks using a terminal.
66 tasks
terminal-bench-science
Research workflows across the life, physical, earth, and mathematical sciences.
70 tasks
terminal-bench
Version 2.1 of Terminal-Bench, evaluating agents in terminal environments.
89 tasks
datacurve
Frontier coding agents on original, long-horizon engineering tasks.
113 tasks
abundant
Ultra long-horizon software engineering tasks from SWE-Marathon.
20 tasks
scale-ai
Deep codebase comprehension and question answering across real repositories.
124 tasks