246 Next AI Use Case Frontier – Protein Sequencing = Model Size +290% Annually to 98 Billion Parameters Over Four Years Note: List of models may not be comprehensive. Source: Stanford RAISE Health via Nestor Maslej et al., ‘The AI Index 2025 Annual Report,’ AI Index Steering Committee, Stanford HAI (4/25) Next AI Use Case Frontiers = Broad & Varied Per Stanford HAI (4/25): The past year has witnessed remarkable progress in AI models applied to protein sequences. Large-scale machine learning models have improved our ability to predict protein properties, accelerating research in structural biology and molecular engineering…These AI-driven approaches have transformed protein science by minimizing reliance on costly, time-intensive experimental methods, enabling rapid exploration of protein function and design. Size of Major Protein Sequencing Models (B Parameters) – 2020-2024, per Stanford RAISE Health 0 50 100 ProGen ProtBert ProGen 2 ProT5 ESM2 ESM3 +290% / Year Number of Parameters, B 2020 2022 2023 2024