Official Sites/Stores:
Affiliates:
The Snapdragon 8797 is designed to push the boundaries of what is possible within a vehicle's electrical/electronic (EE) architecture. Centralized Processing & AI Power
The problem was the ghost.
, scheduled for the first half of 2026, is the first vehicle globally to launch with dual Snapdragon 8797 chips qualcomm 8797
For autonomous driving, the 8797 (under the Snapdragon Ride portfolio) offers superior AI performance for perception, planning, and control tasks. It provides the necessary compute power to handle multiple sensors simultaneously, improving safety and reliability. 3. Software-Defined Vehicle (SDV) Architecture
: It features a peak computing power of up to 640 TOPS (Tera Operations Per Second), directly competing with NVIDIA's Thor series. The Snapdragon 8797 is designed to push the
A key feature of the Qualcomm 8797 is its dual-purpose capability as a central domain controller , designed to power both smart cockpits autonomous driving systems simultaneously within a single platform. 电子工程世界（EEWorld）
| Category | Specification | |---|---| | | TSMC 4nm N4 | | CPU Architecture | Qualcomm Oryon (custom architecture) | | CPU Compute Power | Up to 660K DMIPS | | GPU Compute Power | Up to 8.1 TFLOPS (FP32) | | AI Compute Power | Up to 320 TOPS (dense) / 640 TOPS (sparse) | | Memory Support | Up to 64GB LPDDR5X | | Maximum Displays Supported | 8 screens (including 3K/4K high-definition displays) | | Peripheral I/O | Supports up to 40+ cameras and multi-modal sensors | | Package Form | QAM8797P module, 2253-pin FCBGA+HS package | | Application Domains | Intelligent cockpit, ADAS, cabin-driving fusion, VLA large models | It provides the necessary compute power to handle
: The Neural Processing Unit (NPU) for AI workloads has seen the most dramatic upgrade, with a 12-fold increase in AI performance . This is designed for multi-modal AI, enabling the chip to handle large language models (LLMs) and vision-language-action (VLA) models natively on the device. The raw AI compute power is rated at 320 TOPS (Tera Operations Per Second) , which, due to Qualcomm's use of dense compute metrics, is equivalent to roughly 640 TOPS of sparse compute capacity commonly referenced in the industry.
The architecture of the Qualcomm 8797 is engineered for raw processing power, real-time response, and maximum energy efficiency:
The Qualcomm 8797 represents a massive leap in processing power and energy efficiency. It is built to serve as a long-term hardware foundation capable of supporting automotive software cycles for the next five years.
: To handle complex visual data and artificial intelligence, the on-board Neural Processing Unit (NPU) delivers a staggering 12-fold increase in AI performance compared to previous-generation automotive chips.