Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Reasoning benchmark
BioBasicsBench: everyday wet-lab science. Every model the catalog carries a published BioBasicsBench value for, ranked by that value.
OpenAI and Ligo's benchmark of how well AI agents interpret raw wet-lab evidence, such as gel and blot images, colony plates, microscopy, sequencing files and instrument exports, spot experimental errors and recognize when the data cannot support a conclusion. 7 models in the ModelChoice catalog have a published value.
Each point is a model with a published BioBasicsBench value and complete token prices, on the 75% input / 25% output blend. Frontier models are spaced evenly along the cost axis.
7 models with a published BioBasicsBench value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Mean reward (% of rubric credit), five attempts per task |
|---|---|---|---|
| 1 | OpenAI | 67.0 | |
| 2 | OpenAI | 62.6 | |
| 3 | OpenAI | 45.6 |
What the benchmark measures and how its value is published, from the source’s own record.
BioBasicsBench (BioBasicsBench: everyday wet-lab science) is a reasoning benchmark published by Measuring how AI agents handle everyday wet-lab science. OpenAI and Ligo's benchmark of how well AI agents interpret raw wet-lab evidence, such as gel and blot images, colony plates, microscopy, sequencing files and instrument exports, spot experimental errors and recognize when the data cannot support a conclusion. Its evaluation set is described by the source as 100 routine wet-lab analysis tasks (88 multimodal) across five domains. The source rates its difficulty as routine wet-lab data interpretation.
Each model's result is published by Measuring how AI agents handle everyday wet-lab science as mean reward (% of rubric credit), five attempts per task. ModelChoice prints that value to 1 decimal place, exactly as published, and never converts or rescales it. This page ranks the 7 models that carry a published value, highest value first; a model the source never scored is left out rather than ranked at zero. The source does not state whether a higher value is the better result, so neither does this page.
| 4 | OpenAI | 37.6 |
| 5 | OpenAI | 35.4 |
| 6 | OpenAI | 33.2 |
| 7 | OpenAI | 31.2 |
Evidence key: Observed
Rows are ordered by the value Measuring how AI agents handle everyday wet-lab science published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.