
ICORE-3576JQ38 Industrial 38mm System On Module
iCore-3576JQ38 Mini SoM: A compact 38Ă—38mm powerhouse delivering 6 TOPS edge AI performance, enabling private large-model deployment with industrial-grade reliability for intelligent edge solutions. Especially suitable for IPC Camera and drone applications.
- Product Datasheet: Click Here
- Product Wiki: Click Here
Feature
- 38mm Ă— 38mm Mini Size: The SoM utilizes a BTB interface design and features a high-speed, industrial-grade connector. This design offers superior transmission capabilities, high-frequency stability, and solder-free operation. Measuring just 38 x 38 mm, the SoM saves valuable space.
- AIoT Processor RK3576: The RK3576 utilizes the big.LITTLE architecture (4x A72 + 4x A53) with a clock speed of up to 2.2GHz, providing strong support for high-performance computing and multi-tasking. Equipped with a Mali-G52 MC3 GPU with 145G FLOPS of GPU performance, it enables efficient heterogeneous computing and meets the needs of graphics-intensive applications.
- 6 TOPS Edge Computing Power: Built-in 6TOPS NPU, supports INT4/INT8/INT16/FP16/BF16/TF32 operations, dual-core collaborative or independent operation, and supports parallel multi-tasking and multi-scenario processing.
- Deployment of Large AI Models: Supports private deployment of large-parameter models based on the Transformer architecture, such as the Gemma series, ChatGLM series, and other large language models.
- 8K/4K High-Frame-Rate Video Decoding: Supports 8K@30fps/4K@120fps (VP9, AVS2, AV1) and 4K@60fps video decoding (H.264/AVC), as well as 4K@60fps video encoding (H.264/AVC).
Hardware Overview

More Images



ICORE-3576JQ38 Industrial 38mm System On Module
iCore-3576JQ38 Mini SoM: A compact 38Ă—38mm powerhouse delivering 6 TOPS edge AI performance, enabling private large-model deployment with industrial-grade reliability for intelligent edge solutions. Especially suitable for IPC Camera and drone applications.
- Product Datasheet: Click Here
- Product Wiki: Click Here
Feature
- 38mm Ă— 38mm Mini Size: The SoM utilizes a BTB interface design and features a high-speed, industrial-grade connector. This design offers superior transmission capabilities, high-frequency stability, and solder-free operation. Measuring just 38 x 38 mm, the SoM saves valuable space.
- AIoT Processor RK3576: The RK3576 utilizes the big.LITTLE architecture (4x A72 + 4x A53) with a clock speed of up to 2.2GHz, providing strong support for high-performance computing and multi-tasking. Equipped with a Mali-G52 MC3 GPU with 145G FLOPS of GPU performance, it enables efficient heterogeneous computing and meets the needs of graphics-intensive applications.
- 6 TOPS Edge Computing Power: Built-in 6TOPS NPU, supports INT4/INT8/INT16/FP16/BF16/TF32 operations, dual-core collaborative or independent operation, and supports parallel multi-tasking and multi-scenario processing.
- Deployment of Large AI Models: Supports private deployment of large-parameter models based on the Transformer architecture, such as the Gemma series, ChatGLM series, and other large language models.
- 8K/4K High-Frame-Rate Video Decoding: Supports 8K@30fps/4K@120fps (VP9, AVS2, AV1) and 4K@60fps video decoding (H.264/AVC), as well as 4K@60fps video encoding (H.264/AVC).
Hardware Overview

Product Information
Product Information
Shipping & Returns
Shipping & Returns
Description
iCore-3576JQ38 Mini SoM: A compact 38Ă—38mm powerhouse delivering 6 TOPS edge AI performance, enabling private large-model deployment with industrial-grade reliability for intelligent edge solutions. Especially suitable for IPC Camera and drone applications.
- Product Datasheet: Click Here
- Product Wiki: Click Here
Feature
- 38mm Ă— 38mm Mini Size: The SoM utilizes a BTB interface design and features a high-speed, industrial-grade connector. This design offers superior transmission capabilities, high-frequency stability, and solder-free operation. Measuring just 38 x 38 mm, the SoM saves valuable space.
- AIoT Processor RK3576: The RK3576 utilizes the big.LITTLE architecture (4x A72 + 4x A53) with a clock speed of up to 2.2GHz, providing strong support for high-performance computing and multi-tasking. Equipped with a Mali-G52 MC3 GPU with 145G FLOPS of GPU performance, it enables efficient heterogeneous computing and meets the needs of graphics-intensive applications.
- 6 TOPS Edge Computing Power: Built-in 6TOPS NPU, supports INT4/INT8/INT16/FP16/BF16/TF32 operations, dual-core collaborative or independent operation, and supports parallel multi-tasking and multi-scenario processing.
- Deployment of Large AI Models: Supports private deployment of large-parameter models based on the Transformer architecture, such as the Gemma series, ChatGLM series, and other large language models.
- 8K/4K High-Frame-Rate Video Decoding: Supports 8K@30fps/4K@120fps (VP9, AVS2, AV1) and 4K@60fps video decoding (H.264/AVC), as well as 4K@60fps video encoding (H.264/AVC).
Hardware Overview

















