futurehouse/labbench
Science
Biology
Knowledge
LAB-Bench (Language Agent Biology Benchmark): questions across 8 categories (literature QA, database lookup, sequence manipulation, figure/table reasoning, protocols) testing LLMs on biology-research tasks.
Run this task
CLI:
inspect eval inspect_harbor/futurehouse_labbench --model openai/gpt-5Python:
from inspect_ai import eval
from inspect_harbor import futurehouse_labbench
eval(futurehouse_labbench(), model="openai/gpt-5")Dataset information
| Harbor registry | futurehouse/labbench |
| Inspect task | futurehouse_labbench |
| Latest digest | sha256:0d3d53621c4b583f515cc0b301b030c3835f2c980ca5b58dbdcb1ba42b53e9b7 |
| Samples | 181 |
| Paper | arxiv |
| Source | https://github.com/Future-House/LAB-Bench |
See Task Parameters for the parameter set shared across all Harbor tasks.