futurehouse/bixbench
Science
Biology
Coding
BixBench: real-world bioinformatics analysis capsules with open-answer questions evaluating LLM agents’ ability to author multi-step Jupyter notebooks for biological data analysis.
Run this task
CLI:
inspect eval inspect_harbor/futurehouse_bixbench --model openai/gpt-5Python:
from inspect_ai import eval
from inspect_harbor import futurehouse_bixbench
eval(futurehouse_bixbench(), model="openai/gpt-5")Dataset information
| Harbor registry | futurehouse/bixbench |
| Inspect task | futurehouse_bixbench |
| Latest digest | sha256:2bcaa794dfb1fac3df7b7a4be3e23a09bcec88e86443b843aab095017db2e9f3 |
| Samples | 205 |
| Paper | arxiv |
| Source | https://github.com/Future-House/BixBench |
See Task Parameters for the parameter set shared across all Harbor tasks.