Eyeq4 Datasheet ((full)) ❲CERTIFIED❳
| Function | Benefit | |----------|---------| | 5–8 camera fusion | Reduces blind spots, enables 360° perception | | Hardware CNN engine | Runs semantic segmentation + object detection without choking the CPU | | Internal ISP + HDR | Works with 1 MP–8 MP sensors without external ISP |
The EyeQ4 is fabricated using advanced semiconductor technologies engineered specifically for the thermal and durability demands of the automotive ecosystem. Specification Details 28nm Fully Depleted Silicon On Insulator (FD-SOI) Deep Learning Compute 2.5 TOPS (Tera Operations Per Second) Typical SoC Power Package Type Flip-Chip FBGA 784-pin Package Physical Dimensions 22.5 mm × 22.5 mm × 1.7 mm Ball Pitch ASIL Certification Targets up to ASIL-B / ASIL-D system implementation Fabrications and Power Efficiency Benefits
The EyeQ4 datasheet highlights several next-generation ADAS capabilities:
A thorough datasheet provides block diagrams and timing charts. Here is what you can expect in the perception section: eyeq4 datasheet
: Implements 6 VMP cores . These function as Very Long Instruction Word (VLIW) and Single Instruction Multiple Data (SIMD) vector machines optimized for short integral datatypes common in vision algorithms and neural network operations.
Four CPU cores, each featuring four hardware threads. Performance Metrics
Deep Learning Accelerator. Dedicated high-performance AI engine. The main source of horse power for convolutional neural networks. ZF and Mobileye Safety Technology Chosen by Toyota | Function | Benefit | |----------|---------| | 5–8
The following data summarizes the key specifications and architectural details typically found in an . Core Performance Specifications
Supports Mobileye’s crowdsourced mapping technology, allowing vehicles to create high-definition maps for navigation.
Two cores more versatile than GPUs and more efficient than CPUs. PMA (Programmable Macro Array): These function as Very Long Instruction Word (VLIW)
28nm Fully Depleted Silicon On Insulator (FD-SOI) by STMicroelectronics Up to 8 cameras simultaneously at 36 FPS Safety Standard ISO 26262 compliant with ASIL-B(D) safety level Packaging Flip-Chip FBGA 784-pin; 22.5 x 22.5 x 1.7 mm EyeQ4 Variant Differences
: Multi-threaded execution modules optimized to perform specific pipeline steps more efficiently than standard CPUs, while maintaining greater flexibility than rigid ASICs.
The EyeQ4 architecture utilizes a heterogeneous mix of specialized accelerators to achieve high efficiency. Specification 2.5 TOPS (High variant) / ~1.1 TOPS (Mid variant) Power Consumption ~3 Watts (Automotive grade) CPU Cores 4 multi-threaded MIPS InterAptiv cores (4 threads each) Vision Accelerators
: Includes 1× parallel video image preprocessing port for legacy sensor arrays or secondary inputs.
