Local QCS8550 compute
Six-core (AA) or eight-core (AC) Kryo CPU with Adreno 740 and AI compute for local inference and multimedia processing.
Classic development board
Bring multimedia and AI together at the edge
Powered by QCS8550 in six-core (AA) and eight-core (AC) configurations. Local inference, USB and networking provide a development platform for multimedia interaction, edge video and robotics.

48 TOPS
Maximum AI performance (INT8, dense)
16 GB + 128 GB
LPDDR5X / UFS
3.2 GHz
Maximum six-core / eight-core CPU frequency
4 × USB-A
USB 3.0 peripheral connectivity
Match the compute platform to your device tasks, models, peripherals and product form factor.
Six-core (AA) or eight-core (AC) Kryo CPU with Adreno 740 and AI compute for local inference and multimedia processing.
Supports AidLux (Android 13 + Ubuntu 22.04) or Ubuntu 22.04; match the software option to the delivered version.
HDMI input / output and USB-C with DP 1.4 support capture, preview and interactive prototyping.
Connect external devices through LAN / WAN, USB and 40-pin expansion; confirm driver and peripheral compatibility for the project.
Locate connectors on the hardware and plan device connections around your peripherals.

Select a number or interface name for connection details. Refer to the specification for pin assignments, voltage levels and multiplexing.
DC 12 V, typically a 12 V / 5 A supply. Connect power through DC_IN.
Choose device workloads and integration paths based on platform capabilities and this model's interfaces.
Qualcomm Dragonwing QCS8550

Combine cameras, ISP and local inference for more complex visual models and multitasking in intelligent vision devices.

Connect voice, vision and local models to explore conversation, content understanding and task assistance.

Connect cameras, networks and existing equipment to run recognition, analytics and intelligent applications in a compact compute device.

Combine content, models and application interfaces with local multimodal capabilities in display-equipped interactive products.
Device images illustrate application directions; peripherals, enclosure design and software require project-specific integration.
Configurations below follow the classic development board documentation. Module capabilities and interfaces physically exposed on the complete board are distinguished.
| Compute platform | Qualcomm Dragonwing QCS8550 |
|---|---|
| Process | 4 nm |
| CPU | Qualcomm Kryo, six-core (AA) / eight-core (AC), up to 3.2 GHz; according to the order configuration |
| NPU | Qualcomm Hexagon |
| AI performance (INT8, dense) | Up to 48 TOPS(INT8, dense) |
| AI performance (FP16) | Up to 12 TOPS(FP16) |
| GPU | Qualcomm Adreno 740,680 MHz |
| Memory | 16 GB LPDDR5X |
| Storage | 128 GB UFS |
| Cameras | 3 × 4-lane D-PHY MIPI CSI,22-pin FPC |
|---|---|
| HDMI output | 1 HDMI_OUT below the front-left USB-A group; confirm modes and frame rates against the BSP |
| HDMI input | 1 HDMI_IN below the front-right USB-A group; capture formats depend on the companion driver |
| Video processing | Platform maximum: 8K60 / 4K240 decoding; 8K30 / 4K120 encoding |
| Other display options | USB-C supports DP 1.4. MIPI DSI is listed in the overview but not clearly located in the current drawing; obtain the companion definition before connecting a display. |
| USB | 4 × USB-A 3.0 Host + 1 × USB-C 3.0 OTG / DP |
|---|---|
| Network | 4 × Gigabit Ethernet RJ45 via QCA8337 switch; default 1 WAN + 3 LAN |
| Wireless | Wi-Fi 7(802.11be,2.4/5/6 GHz,2 × 2 MIMO);Bluetooth 5.3 |
| Audio | 3.5 mm Line-in / Line-out |
| Expansion | 2 groups of 40-pin headers named 3.3 V and 1.8 V GPIO; refer to the pin table for individual voltages, multiplexing and power |
| SSD expansion | Rear M.2 M-Key 2280, PCIe Gen 3 ×2, supports NVMe SSDs |
| Removable storage | Rear Micro SD / TF slot |
| Operating System | AidLux (Android 13 + Ubuntu 22.04) or Ubuntu 22.04 |
|---|---|
| Power | DC 12 V, typically a 12 V / 5 A supply |
| Operating temperature | -20 °C~+50 °C |
| Dimensions | 125 × 110 × 35.5 mm |
| RTC | RTC backup battery connection; confirm battery type, polarity and installation for the purchased board variant |
| Fan | 4-pin: 5 V, PWM INPUT, reserved, GND |
AI figures are platform-rated peaks. Dense, sparse and FP16 figures are listed separately; actual performance depends on the model, precision, software and cooling. Input power ratings do not represent total board consumption. Accessories and system images depend on the delivered configuration.
Six-core (AA) and eight-core (AC) variants depend on the order configuration. The system image must match the specific variant.
Download Chinese and English product briefs for configurations and application directions. Obtain detailed specifications for carrier board and device integration.
Product overview, key capabilities and application directions.
PDF · ChineseDownload Chinese briefProduct overview, key capabilities and application directions.
PDF · EnglishDownload English briefSystem images, firmware, development guides and reference designs.
Share your model, peripheral, mechanical and system requirements to define the product and delivery configuration together.