RoboticsTechNodeabout 6 hours ago
XPENG releases TuringViT for driving and robots
1 min read
XPENG published TuringViT, a vision encoder for smart driving, cockpit systems, and IRON humanoid robots. Two variants, TuringViT-18L and TuringViT-24L, were released. At 1536x1536 resolution, TuringViT-18L achieved 3.04 times the throughput of Seed1.5-ViT and 2.16 times SigLIP2-ViT-L. The model trained on 850 million image-text pairs scored 83.6% across six zero-shot benchmarks.
Level
Hype check
Tap to vote and see what everyone thinks.
