cohere/terminal_bench_aai
Coding
Assistants
AAII Terminal-Bench Hard: 44 hard tasks from terminal-bench v1 (terminal-bench-core==0.1.1), used with the Terminus-2 agent.
Run this task
CLI:
inspect eval inspect_harbor/cohere_terminal_bench_aai --model openai/gpt-5Python:
from inspect_ai import eval
from inspect_harbor import cohere_terminal_bench_aai
eval(cohere_terminal_bench_aai(), model="openai/gpt-5")Dataset information
| Harbor registry | cohere/terminal_bench_aai |
| Inspect task | cohere_terminal_bench_aai |
| Latest digest | sha256:d8908945101bb9a713690ac46539d13ca7d44658637f2650a857470926c191b9 |
| Samples | 44 |
| Paper | arxiv |
| Source | https://github.com/harbor-framework/terminal-bench |
See Task Parameters for the parameter set shared across all Harbor tasks.