labbench@1.0
Science
Biology
Knowledge
LAB-Bench FigQA: 181 scientific figure reasoning tasks in biology from Future-House LAB-Bench.
Run this task
CLI:
inspect eval inspect_harbor/labbench_1_0 --model openai/gpt-5Python:
from inspect_ai import eval
from inspect_harbor import labbench_1_0
eval(labbench_1_0(), model="openai/gpt-5")Dataset information
| Harbor registry | labbench@1.0 |
| Inspect task | labbench_1_0 |
| Version | 1.0 |
| Samples | 181 |
| Paper | arxiv |
See Task Parameters for the parameter set shared across all Harbor tasks.