HW3 is based on a custom Tesla-designed system on a chip called “FSD Chip”, fabricated using a 14 nm process by Samsung. Jim Keller and Pete Bannon, among other architects, have led the project since February 2016. FSD Chip features twelve ARM Cortex-A72 CPUs operating at 2.6 GHz, two systolic arrays (not unlike the approach of TPU) operating at 2 GHz and a Mali GPU operating at 1 GHz. Tesla claimed that FSD Chip processes images at 2,300 frames per second (fps), which is a 21× improvement over the 110 fps image processing capability of HW2.5. The firm described FSD Chip as a “neural network accelerator” custom-designed for Tesla AI processing. Each of the two systolic arrays on a single FSD Chip are capable of 36 trillion operations per second, and there are two FSD Chips for redundancy.
HW3 cars are equipped with eight cameras, in the same locations and covering the same directions as the HW2 and HW2.5 cameras. Each of the eight cameras supplied with HW3 use the same AR0136A image sensor supplied by Onsemi, which has a maximum resolution of 1280×960 (1.2-megapixel) and a 3.75 μm pixel size. Initial versions of HW3 also included a Continental ARS4-B radar module.
https://en.wikipedia.org/wiki/Tesla_Autopilot_hardware
Samsung manufactures Tesla’s AI4 chips, which power the company’s Full Self-Driving (FSD) hardware (also known as Hardware 4 or HW4). These custom system-on-chip (SoC) processors are produced on a 7 nm process at Samsung’s facility in Hwaseong, South Korea. Tesla designs the chips internally, but Samsung handles the fabrication. This partnership was confirmed by Tesla CEO Elon Musk and multiple reports, including a recent $16.5 billion multi-year deal for future chips that builds on Samsung’s existing role with AI4.