Insilico Medicine Releases Open Longevity AI Toolkit in Cell Study

Insilico Medicine published in Cell an open toolkit for AI in aging biology: the LongevityBench benchmark of 17 tasks, five compact Longevity-LLM models (0.6–9 billion parameters), and the Longevity Claw agent platform. The compact models outperformed frontier systems: L-Qwen3.5-9B showed the best result, and L-Qwen3-0.6B had a 5.7-year error in proteomic age prediction versus 10.1 for the best frontier model.

• LongevityBench includes 17 tasks in five domains: clinical, genetics, epigenetics, transcriptomics, proteomics
• L-Qwen3.5-9B is the best model with an aggregate score of 4.4 versus 8.2 for Gemini 3.1 Pro
• Longevity Claw proposed 328 candidate genes for aging intervention
• Models and data are on Hugging Face, platform code on GitHub under MIT license