The CORE-1126BJD4 is equipped with Rockchip's quad-core RV1126B AI vision processor, an integrated 3 TOPS NPU, and supports lightweight AI models and deep learning frameworks. It also features a built-in 12MP ISP, an 8MP AI-ISP, and multiple algorithm accelerators. This SoM supports 12MP video encoding, 4K video decoding, and simultaneous input from four cameras.

• Product Datasheet: Click Here

Feature

• AI Vision Processor RV1126B: Equipped with a quad-core 64-bit RV1126B AI vision processor. Based on the ARM Cortex-A53 core, it runs at up to 1.6 GHz and includes NEON advanced SIMD and an FPU, enhancing high-performance computing and multitasking capabilities.
• Deployment of Lightweight AI models: Supports lightweight large language models, large visual models, or multimodal large models with less than 2B parameters, such as Qwen, Gemma2-2B, Phi2, MiniCPM series, and other small AI large models.
• Built-in 12MP ISP: Equipped with a new generation of hardware ISP, it supports up to 12 megapixels and features key algorithm accelerators, including HDR, 3A, LSC, and noise reduction. On this basis, it also adds AI-ISP (supports up to 8 megapixels).
• 4K High-Frame-Rate Video Decoding: Supports video encoding up to 12MP at 30fps. Its multi-stream encoding capability allows simultaneous local high-resolution recording and low-resolution streaming to the cloud. It supports 4K at 30fps decoding and integrates the JPEG codec with Baseline mode for various image sizes and YUV formats.
• Simultaneous Input From 4 Cameras: It supports two MIPI CSI/LVDS/SubLVDS D-PHY interfaces and one DVP (BT.601/BT.656/BT.1120) interface, allowing up to four cameras to input simultaneously.

Hardware Overview