From the Lab
Thinking Out Loud
Research notes and perspectives from the Primate Intelligence team.
Research
Darwin: Video JEPA model that outperforms SOTA models while running on edge CPU
Language models had their GPT moment. Vision, especially video, hasn’t yet!
ResearchBenchmarks Deep Dive — Coming Pre-Launch
Our full benchmark methodology, numbers against YOLO / SAM / VLM baselines, and the rig we ran them on. In production — publishing ahead of public launch.
ResearchHow JEPA Works — Coming Pre-Launch
Why we build on Joint-Embedding Predictive Architectures instead of pixel-space reconstruction. The deep dive is in production — publishing ahead of public launch.
Subscribe via RSS. Every post is also available as raw markdown — append .md to any post URL.