OpenAI Adds Life Sciences Model to API, Targets Pharma Drug Discovery

On October 10, OpenAI added an unannounced model, gpt-rosalind-discovery, to its API pricing page under a newly created Life Sciences Models category. Priced at $5 per million input tokens and $25 per million output tokens, the model joins the Rosalind platform, OpenAI’s dedicated life sciences AI brand. With global API access restricted to vetted research institutions and pharmaceutical companies, this deployment marks a strategic departure from general-purpose models, establishing the first standalone, continuously vetted artificial intelligence line explicitly engineered for drug discovery among major foundation model providers. The Rosalind initiative launched in April as a U.S.-focused research preview for enterprise clients, later expanding globally and integrating with scientific tooling. OpenAI has secured partnerships with Amgen, Moderna, Eli Lilly, and Novartis, while collaborating with Ginkgo Bioworks to automate laboratory workflows, successfully reducing protein synthesis costs by 39 percent. The company explicitly positions itself as a B2B tool supplier, contrasting with competitors like Anthropic that develop drug pipelines internally. By restricting model access to approved institutions and prohibiting customer-facing product integration, OpenAI avoids direct pharmaceutical competition while capturing high-value enterprise licensing revenue. The life sciences sector represents a critical artificial intelligence growth frontier, having attracted over $17 billion in cumulative funding since 2019. Commercialization, however, remains constrained by the inherent latency of wet-lab validation and clinical trials, which slow rapid model iteration. Current systems primarily automate target identification and experimental design, leaving strategic direction to human researchers. Splitting Rosalind into research and discovery tiers reflects an industry shift toward modular AI specialization, enabling targeted optimization for discrete R&D workflows. Maintaining identical pricing across both models indicates a strategy prioritizing ecosystem adoption and real-world usage data over premium monetization. The success of this architecture will ultimately depend on OpenAI’s ability to demonstrate measurable efficiency gains and accelerate the feedback loop between computational prediction and laboratory validation.