GPT-Rosalind: Frontier Model for Life Sciences Research
OpenAI describes GPT-Rosalind as its frontier reasoning model series built for life sciences research, helping researchers accelerate early discovery across biology, drug discovery and translational medicine; an update on 11 September 2026 says it has left research preview and is available globally to eligible organizations through OpenAI's trusted-access program, with published pricing taking effect on 5 October 2026
Tool Overview
Features, steps and FAQ below
Features
- ✓ Scientific focus: OpenAI says it is built for multi-step research workflows such as literature review, sequence-to-function interpretation, experimental planning and data analysis, and delivers its best performance on tasks requiring reasoning over molecules, proteins, genes, pathways and disease-relevant biology
- ✓ Tools and databases: OpenAI says it is more effective at selecting and using the right computational tools, databases and domain capabilities in multi-step workflows, and it released a freely accessible Life Sciences research plugin for Codex on GitHub that connects scientists to over 50 scientific tools and data sources
- ✓ Benchmarks: OpenAI says it achieved leading performance among models with published scores on BixBench, built around real-world bioinformatics and data analysis, and outperforms GPT-5.4 on 6 of 11 tasks on LABBench2, with the biggest gain on CloningQA, which requires end-to-end design of DNA and enzyme reagents
- ✓ Third-party evaluation: OpenAI says it partnered with Dyno Therapeutics on an RNA sequence-to-function prediction and generation task using unpublished, uncontaminated sequences, where best-of-ten model submissions in the Codex app ranked above the 95th percentile of human experts on prediction and around the 84th percentile on generation
- ✓ Access and pricing: OpenAI says GPT-Rosalind is available in ChatGPT, Codex and the API to qualified customers through its trusted access program, limited to approved internal research and not for customer-facing products or external commercial applications, with published pricing taking effect on 5 October 2026 and cache-write pricing not applying
How to Use
- Apply for access through OpenAI's trusted-access program (for eligible life sciences organizations)
- Call it in ChatGPT, Codex or the API as gpt-rosalind-research, keeping use to approved internal research
- Install the official Life Sciences research plugin from GitHub to connect over 50 scientific tools and data sources
- Pilot it on literature review, experimental planning and data analysis workflows and validate it on your own tasks
FAQ
What is GPT-Rosalind?
OpenAI describes it as its frontier reasoning model series built to support research across biology, drug discovery and translational medicine, named after Rosalind Franklin.
What stage is it at?
An OpenAI update on 11 September 2026 says it is coming out of research preview and is now available globally to eligible organizations through the trusted-access program, with published pricing taking effect on 5 October 2026.
Can anyone use it?
OpenAI says access requires approval through its trusted-access program and is limited to approved internal research; it is not available for customer-facing products or external commercial applications at this time.
What tooling ships with it?
OpenAI says it also offers a freely accessible Life Sciences research plugin for Codex on GitHub, including modular skills for human genetics, functional genomics, protein structure, biochemistry, clinical evidence and public study discovery that connect to over 50 scientific tools and data sources.
Is there third-party validation?
OpenAI says it partnered with Dyno Therapeutics on an RNA sequence-to-function task using unpublished, uncontaminated sequences, where best-of-ten submissions in the Codex app ranked above the 95th percentile of human experts on prediction and around the 84th percentile on generation, and that it achieved leading published performance on BixBench.