Self-Driving Supercomputer Showdown: NVIDIA Drive Thor vs Tesla FSD Hardware 4 vs Qualcomm Snapdragon Ride Flex
NVIDIA Drive Thor delivers up to 1,000 TOPS (INT8 sparse) on a single chip and can be paired via NVLink-C2C to about 2,000 TOPS for high-end Level 4–5 autonomy. Thor combines a Blackwell GPU with 2560 CUDA cores and 96 Tensor Cores, a 14-core Arm Neoverse V3 CPU complex, on a 4nm TSMC process, and supports up to 128 GB LPDDR5X memory. Thor is designed as a centralized automotive computer unifying autonomous driving, ADAS, digital cockpit and infotainment on one platform. Production of Drive Thor is slated to begin in 2025, with design wins at Li Auto, ZEEKR, BYD, Hyper