BioBasics Bench
An evaluation benchmark targeting foundational scientific capabilities, distinguishing models' levels of scientific literacy through tasks such as multi-step reasoning, chart interpretation, and experimental design.
Timeline (last 90 days)
OpenAI's official website published a benchmark called BioBasics Bench, used to assess a model's ability to understand and analyze everyday biology experiment materials, but the body text explicitly states that only 7 models were tested, and 6.2 Astra was not among them.
In the official chart, the previous generation GPT-6 Astra scored an average of 66.7% on BioBasics Bench (the hover tooltip shows about 67%).
In a secondhand screenshot circulating online, the entry labeled 6.2 Astra shows a score of 89.0%, but the source, accuracy and whether the webpage was modified cannot be verified.
BioBasics Bench is a domain-specific benchmark, and unlike general evaluations such as MMLU and GPQA, it specifically assesses a model's reading comprehension and analytical ability on everyday biology experiment materials.
The leaked BioBasics Bench data mentioned above has not yet been independently verified, and the related vulnerability is reportedly already patched.
OpenAI, together with a biotech company, released a new benchmark, BioBasics Bench, containing 100 everyday lab tasks, requiring agents to draw conclusions from raw readings; the leaderboard and description have been made public.
All Facts (6)
In the official chart, the previous generation GPT-6 Astra scored an average of 66.7% on BioBasics Bench (the hover tooltip shows about 67%).
60%UnverifiedOpenAI's official website published a benchmark called BioBasics Bench, used to assess a model's ability to understand and analyze everyday biology experiment materials, but the body text explicitly states that only 7 models were tested, and 6.2 Astra was not among them.
50%UnverifiedIn a secondhand screenshot circulating online, the entry labeled 6.2 Astra shows a score of 89.0%, but the source, accuracy and whether the webpage was modified cannot be verified.
50%UnverifiedBioBasics Bench is a domain-specific benchmark, and unlike general evaluations such as MMLU and GPQA, it specifically assesses a model's reading comprehension and analytical ability on everyday biology experiment materials.
50%UnverifiedThe leaked BioBasics Bench data mentioned above has not yet been independently verified, and the related vulnerability is reportedly already patched.
50%UnverifiedOpenAI, together with a biotech company, released a new benchmark, BioBasics Bench, containing 100 everyday lab tasks, requiring agents to draw conclusions from raw readings; the leaderboard and description have been made public.
50%