OrangePi O1 is powered by BM1688, 8-core ARM Cortex-A53 @ 1650Mhz main processor, RISCV C906 @ 1200Mhz co-processor, and built-in SOPHGO neural network acceleration engine TPU, with computing power up to 32TOPS@INT4, 16TOPS@INT8, 4TOPS@FP16/ BF16, 0.5TOPS/@FP32. With 8GB/16GB LPDDR4/LPDDR4X memory, external 16GB/32GB/64GB/128GB eMMC module; onboard M.2 interface, supports access to NVMe SSD hard disk to meet the needs of fast read/write and high-capacity storage; supports the expansion of 4G/5G/WiFi6E+ BT 5.3, which facilitates the deployment of edge-oriented services; supports H.264 and H.265 16-channel 1080P @30fps video decoding and H.264 and H.265 10-channel 1080P @30fps video encoding, which meets the more complex and diversified application requirements of scenarios.
OrangePi O1 has a wealth of interfaces, including USB3.0, USB2.0, HDMI, Gigabit Ethernet, MIPI camera interface, TF card slot, Type-C port power supply, 26Pin expansion connector, fan connector, etc., which greatly meets the user's needs to connect a variety of peripherals.
With high performance and low power consumption, OrangePi O1 can realize intelligent applications such as image classification, target detection, instance segmentation, semantic segmentation, behavioral analysis, text recognition, natural language processing, speech recognition, voice synthesis, search recommendation and so on by efficiently adapting the algorithmic model, which is widely used in edge computing, intelligent transportation, intelligent campus, intelligent industrial gateway, intelligent industrial controller, AIoT, in-vehicle platforms and other fields.
Supports 16-channel 1080P video decoding and 10-channel 1080P video encoding, suitable for video surveillance, streaming media processing and other applications. Efficient JPEG image coding and decoding capabilities to meet image processing needs.
Provides a variety of storage options (eMMC, SPI FLASH, NVMe SSD, SATA SSD, uSD card) to enhance the connectivity and expandability of the device.
Supports the private deployment of large-scale parameter models under the Transformer architecture, such as Gemma-2B, Qwen1.5-1.8B, Llama2-7B, ChatGLM3-6B and other large language models.
| SoC | BM1688 |
| CPU | Main Processor: 8-core ARM Cortex-A53 @ 1650Mhz Co-processor: RISC-V C906 @ 1200Mhz |
| TPU | Built-in SOPHGO neural network acceleration engine TPU with 32T@INT4, 16T@INT8, 4T@FP16/BF16, 0.5T/@FP32. |
| Video decoding | H.264 and H.265:16 channels 1080P @30fps |
| Video encoding | H.264 and H.265: 10 channels 1080P @30fps |
| Image encoding | JPEG: Decode 480 frames/sec @1080P; Encode 480 frames/sec @1080P |
| Memory | LPDDR4/LPDDR4X :8GB/16GB |
| Storage | eMMC module (optional):16GB/32GB/64GB/128GB SPI Flash: 128Mb、256Mb （Choose either eMMC or SPI flash） M.2 M-KEY Socket：PCIe3.0 M.2 NVME SSD SATA Socket：SATA3.0 SSD uSD card slot：supports up to 128GB uSD card |
| Wi-Fi+Bluetooth | Wi-Fi + Bluetooth 2-in-1 module Wi-Fi 6E+BT 5.3, BLE |
| 4G/5G module | M.2 B-KEY Socket |
| Ethernet transceiver | 2* 10/100/1000Mbps Ethernet (onboard PHY chip: YT8531C) |
| Display | 1* HDMI TX 2.0 interface up to 4Kp60 1* 4 lane MIPI DSI via FPC connector |
| Camera | 2* 4-lane MIPI CSI camera interface |
| USB | 2*USB 3.0 HOST 2*USB 2.0 HOST |
| Audio | 3.5mm headphone jack audio input/output 2PIN vertical terminal block speaker output connector Electret Condenser Micphone audio input |
| Pushbutton | 1* On/Off button, 1*RESET button |
| FAN | 2PIN DC12V fan connector |
| RTC | 2PIN battery backup connector |
| 26PIN | 26PIN function expansion interface that supports the following interface types: GPIO, UART, I2C, SPI, PWM |
| Power supply | Type- C 20V/3.25A |
| PCB | 90 x 70 mm |
| Weight | 76g |
