Liquid AI's tiny longevity models top GPT-5 and Gemini in Cell-published biology benchmark
JosephJacks_ · x · 2026-09-18
Liquid AI, partnering with Insilico Medicine, published a Cell cover paper claiming its Liquid Foundation Models set state-of-the-art on biological longevity tasks, outperforming Gemini-3.1-Pro, GPT-5, and Claude Opus.
Releases include:
- LongevityBench: a 17-task eval suite covering clinical records, DNA methylation, transcriptomics, plasma proteomics, and genetic evidence, testing whether general LLMs can interpret aging data.
- LFM2-1.2B-Longevity and LFM2-2.6B-Longevity: compact specialized models that run on personal devices for healthcare and longevity research.
The team argues frontier-class capability on-device is especially valuable in specialized verticals like healthcare. Note the claims come from the vendor's own benchmark suite.
Related event: Insilico and Liquid AI's Longevity LLMs Land Cell Cover, Beat GPT-5(4 posts)→
More from Models
- Open-Weight Model Maker Arcee AI Announces Series B at Over $1B Valuation — code_star · 2026-09-18
- Stanford's 10-person Marin open lab is live-training a 535B model in the open — wandb · 2026-09-18
- Self-Proclaimed ChatGPT Co-Inventor Launches Jev, Claims 200x Speed at 1/400 Cost — iamrobotbear · 2026-09-18
- Codex Pro User Says Usage Limits Got 5-10x Worse, Can't Even Buy Another Plan — Junra · 2026-09-18
- GPT-6 Astra Deciphers an Undeciphered 1918 German WWI Radio Transmission — moultano · 2026-09-18
- Sakana AI Introduces Fugu Max and Fugu Ultra v2 Models — SakanaAILabs · 2026-09-18