Environment-Intensive Task Generation on CLI Terminal Environments
1,655Total InstancesCLI-Gym
Evaluation Results
| Method | Links | |
|---|---|---|
| CLI-GymGold Instance=Pre-installed Github Repo, Data Engine=Agentic Synthesis, Problem Instance=Environment / Failed Unit Tests, Collection=Auto2026.02 | 1,655 | |
| Terminal-Bench@2.0Gold Instance=-, Data Engine=-, Problem Instance=Environment-Intensive Issue +, Collection=Human-written2026.02 | 89 | |
| Terminal-Bench@1.0Gold Instance=-, Data Engine=-, Problem Instance=Environment-Intensive Issue, Collection=Human-written2026.02 | 80 |