Banana Pi BPI-CM5 Pro AI Computer Module Rockchip RK3576

BANANA PI

No reviews yet Write a Review
SKU:
560-BPI-CM5-PRO-4G+32G-NW
MPN:
BPI-CM5-PRO-4G+32G-NW
Banana Pi BPI-CM5 Pro features Rockchip RK3576, 4 GB LPDDR5, 32 GB eMMC, octa-core Arm CPU, Mali-G52 MC3 GPU, 6 TOPS NPU, and advanced 4K/8K video processing. View full description
  • 100% Original Products
  • GST Invoice
  • Earn 3% Cashback
₹23,591.92 inc. GST
₹19,993.15 ex. GST
20 people looking at this.

Technical Specifications

Brand:
BANANA PI
CPU:
RK3576 Quad-core Cortex-A72@2.2GHz and Quad-core Cortex-A53@1.8GHz, 8nm process
GPU:
ARM Mali G52 MC3 GPU, Fully compatible with OpenGL ES 1.1, 2.0, and 3.2, OpenCL up to 2.0, and Vulkan 1.1. Dedicated 2D hardware engine with MMU maximizes display performance and provides very smooth operation.
NPU:
6 TOPS@INT8, Supports INT4, INT8, INT16, FP16, BF16, and TF32 mixed precision, Supports deep learning frameworks: TensorFlow, Caffe, Tflite, Pytorch, Onnx, NNAndroid NN, etc.
VPU/Codec:
Hardware decoding: Supports H.264, H.265, VP9, AV1, and AVS2 up to 8K@30fps or 4K@120fps, and high-quality JPEG decoding up to 4K@60fps., Hardware encoding: Supports H.264 and H.265, JPEG up to 4K@60fps.
ISP:
ISP, 16MP
RAM:
4GB/8GB/16GB (up to 16GB) 32-bit LPDDR5
Flash:
32/64/128GB eMMC, Supports MicroSD card expansion
Display:
1x HDMI OUT 2.1, supports 4K@120fps, 1x MIPI DSI resolution up to 2K@60fps, 1x DP 1.4 resolution up to 4K@120fps
Camera:
1x 4Lanes MIPI CSI, up to 2.5Gbps per lane, 1x 2Lanes MIPI CSI, up to 2.5Gbps per lane
Network:
1x GMAC, provides RGMII/RMII interface, Supports 10/100/1000Mbps data transfer rates
PCIe:
2x PCIe 2.0 1-lane
USB:
1x USB 3.0, 1x USB 2.0
Audio:
SPDIF_Tx x1, AUDIO_R/L x1, i2s x1

Warranty Information

All the products supplied by Evelta are genuine and original. We offer 14 days replacement warranty in case of manufacturing defects. For more details, please visit our cancellation and returns page.

Description

Resources:
Documents

Banana Pi BPI-CM5 Pro AI Compute Module with Rockchip RK3576, 6 TOPS NPU, 4 GB LPDDR5 and 32 GB eMMC

The Banana Pi BPI-CM5 Pro is a compact AIoT compute module powered by the Rockchip RK3576 second-generation 8 nm processor platform. This configuration combines 4 GB LPDDR5 memory with 32 GB eMMC storage and an octa-core CPU architecture consisting of 4 x Cortex-A72 cores operating at up to 2.2 GHz and 4 x Cortex-A53 cores operating at up to 1.8 GHz. A dedicated NEON coprocessor, Arm Mali-G52 MC3 GPU, and integrated NPU delivering up to 6 TOPS of INT8 AI performance make the module suitable for edge AI, machine vision, robotics and industrial embedded applications.

The RK3576 integrates dedicated multimedia engines supporting high-resolution H.264, H.265, VP9, AV1 and AVS2 decoding together with H.264, H.265 and JPEG encoding. The GPU supports OpenGL ES up to 3.2, OpenCL 2.0 and Vulkan 1.1 for graphics and compute acceleration. Measuring only 55 x 40 mm and weighing approximately 12 g, the BPI-CM5 Pro provides substantial compute capability in a compact form factor. A wide 4.5-23 V operating input range and support for multiple operating systems further improve its suitability for industrial controllers, AI gateways, multimedia devices and custom carrier-board designs.

Key Features

  • Rockchip RK3576 SoC: Built around Rockchip's second-generation 8 nm AIoT platform for high-performance embedded computing.
  • Octa-Core CPU Architecture: Integrates 4 x Cortex-A72 cores at up to 2.2 GHz and 4 x Cortex-A53 cores at up to 1.8 GHz.
  • Dedicated NEON Coprocessor: Provides additional acceleration for compatible vector and multimedia processing workloads.
  • 6 TOPS NPU: Integrated AI accelerator delivers up to 6 TOPS of INT8 computing performance.
  • Mixed-Precision AI Support: NPU supports INT4, INT8, and INT16 computation for compatible neural-network workloads.
  • 4 GB LPDDR5 Memory: Provides high-speed system memory for embedded applications, operating systems, and AI workloads.
  • 32 GB eMMC Storage: Integrated non-volatile storage supports operating-system installation, applications, and local data.
  • Arm Mali-G52 MC3 GPU: Provides graphics and compute acceleration for compatible visual and multimedia applications.
  • Advanced Graphics APIs: Supports OpenGL ES 1.1/2.0/3.2, OpenCL 2.0 and Vulkan 1.1.
  • High-Resolution Video Decoding: Supports H.264, H.265, VP9, AV1 and AVS2 decoding up to 8K at 30 fps or 4K at 120 fps, depending on codec and workload.
  • 4K Video Encoding: Supports H.264, H.265, and JPEG encoding up to 4K at 60 fps.
  • Wide Voltage Input: Operates from 4.5 V to 23 V with +/-5% voltage tolerance for flexible embedded-system integration.
  • Compact Compute Module: 55 x 40 mm dimensions simplify integration into space-constrained carrier boards and equipment.
  • Lightweight Design: Approximately 12 g module weight suits compact embedded and mobile systems.
  • Multiple Operating Systems: Designed to support various software platforms for embedded, AI and industrial applications.

Applications

  • AI edge computing
  • Machine vision
  • Intelligent robotics
  • Industrial automation
  • Smart gateways
  • AIoT devices
  • Video analytics
  • Digital signage
  • Multimedia systems
  • Industrial control
  • Edge servers
  • Computer vision systems
  • Smart cameras
  • Embedded Linux systems
  • Custom carrier-board designs
  • AI inference devices

Hardware Interface