GPT-Rosalind, recently released in April 2026, is OpenAI’s model built for life sciences research. It reasons across biology, works with scientific tools, and evaluates evidence to accelerate scientific research and drug discovery.
This model’s capabilities go beyond casual work tasks and personal questions, so advancing its technology is especially crucial. That’s why, about two months after its introduction, GPT-Rosalind is getting new capabilities to further its impact.
New Capabilities for GPT-Rosalind
GPT-Rosalind takes agentic coding and tool-use from GPT-5.5 and combines those skills with stronger model intelligence that specializes in core drug discovery. As a result, GPT-Rosalind now has better performance, reasoning, and workflows.
Improved Performance
To effectively help with research, GPT-Rosalind needs regular improvements. OpenAI created LifeSciBench to measure these improvements, which is a benchmark judged by external life science experts. LifeSciBench measures GPT-Rosalind compared to other AI models using six key categories: evidence handling, analysis, design and optimization, scientific reasoning, validation and operations, and translation and communication.
The overall LifeSciBench scores show that GPT-Rosalind currently outranks other AI models like GPT-5.5, Grok 4.3, and Gemini 3.1 Pro.
Stronger Scientific Reasoning
GPT-Rosalind’s reasoning has improved for medicinal chemistry, quantitative biology, and real-world lab work. For medicinal chemistry, which focuses on turning molecules into beneficial drugs, GPT-Rosalind showed better reasoning and multimodal synthesis in this area than GPT-5.5.
When evaluating GPT-Rosalind’s long horizon analysis for genomics and quantitative biology, it used 31% fewer tokens than GPT-5.5 while still improving accuracy. GPT-Rosalind also proved to be helpful at assisting scientists with lab work, scoring higher than GPT-5.5 with improved token efficiency.
Executing Workflows from Reasoning
GPT-Rosalind can now go beyond reasoning by also assisting with workflows. OpenAI introduced two plugins (Life Sciences Research and Life Sciences NGS Analysis) to extend the model’s intelligence by adding the ability execute repeatable scientific workflows. The plugins add sourced evidence retrieval, biological interpretation, and bioinformatics execution to help scientists more easily connect evidence and analyze data.
Expanded Access
With this new update, GPT-Rosalind is now available for all eligible organizations globally. Eligible organizations must be conducting legitimate scientific research that offers a clear public benefit, has strong governance and safety oversight, and has controlled access with enterprise-grade security.
A great example is Novo Nordisk, who’s scaling medical research to bring innovative treatment options to patients faster. GPT-Rosalind can help with scaling that research.
A Beneficial Tool for Scientific Research Companies
Companies focusing on scientific research to help the public may be able to take advantage of GPT-Rosalind’s capabilities. As this model continues to expand, it can have a positive impact on public health, drug discovery, and translational medicine.
Even if your company isn’t focused on medicine and science, there are plenty of other advanced AI models that can help you grow your brand. Contact Avenue Z today to see how our tailored strategies can help you use growing AI tools to your advantage.
The AI-First Agency
Win AI search, grow revenue and build reputation through PR and digital marketing.
