OpenAI Upgrades GPT-Rosalind with Specialized Biological and Chemistry Reasoning
OpenAI has announced a major update to its specialized GPT-Rosalind model series. GPT-Rosalind is purpose-built for life sciences, drug discovery, and genomics at enterprise scale. Rather than acting as a generalist chatbot, it is engineered to handle scientifically complex workflows, combining GPT-5.5's agentic coding and tool-use capabilities with deep domains of medicinal chemistry, quantitative biology, and wet lab troubleshooting.
To measure GPT-Rosalind's real-world impact, OpenAI evaluated the model using several specialized benchmarks:
- LifeSciBench: An end-to-end, expert-judged benchmark spanning evidence synthesis, experimental design, and wet lab troubleshooting.
- MedChemBench: Evaluates chemical structure understanding, potency prediction, and retrosynthesis. GPT-Rosalind scored 27.5% (compared to GPT-5.5's 25.1%) while using 7.2% fewer tokens.
- GeneBench: Assesses long-horizon genomics and quantitative biology tasks. GPT-Rosalind achieved 21.6% accuracy (vs. GPT-5.5's 20.4%) while using 31% fewer tokens.
- LabWorkBench: Evaluates real-world wet lab protocol troubleshooting. GPT-Rosalind scored 63.2% (vs. GPT-5.5's 55.8%) while using 5.3% fewer tokens.
OpenAI is rolling out the updated model in research preview to eligible global organizations through a "trusted-access deployment structure." OpenAI also announced a strategic partnership with pharmaceutical giant Novo Nordisk to deploy GPT-Rosalind to scale medical research and analyze complex multi-omic datasets.