Anonymous
Apr 16, 2026
Rating: 5/5
Excellent AI Performance
The 6 TOPS NPU handles my object detection models smoothly. Great upgrade from the K210.
Description
Reviews
| Brand | - |
| Category | Electronic Components / Maker DIY / Development Boards & Control Boards |
| Origin | - |
| Processor | Kendryte K230 RISC-V |
| AI Computing Power | 6 TOPS (INT8) |
| RAM | 512MB LPDDR4 |
| Storage | 128MB SPI NAND Flash |
| Camera Interface | MIPI CSI 2-lane |
| Display Interface | MIPI DSI |
| Wireless Connectivity | WiFi 5 + Bluetooth 5.0 |
| Power Input | 5V/2A Type-C |
| Operating System | Linux / CanMV |
| Dimensions | 65mm x 40mm x 10mm |
Anonymous
Apr 16, 2026
Rating: 5/5
Excellent AI Performance
The 6 TOPS NPU handles my object detection models smoothly. Great upgrade from the K210.
Anonymous
Mar 11, 2026
Rating: 4/5
Good board, steep learning curve
Hardware is solid but documentation needs more examples for beginners. Once set up, it works well.
Anonymous
Feb 19, 2026
Rating: 5/5
Perfect for Vision Projects
Integrated camera support is seamless. Using it for a sorting robot project without issues.
Anonymous
Feb 10, 2026
Rating: 5/5
Best in class for price
Unbeatable value for a RISC-V AI board. Python support makes prototyping very fast.
Anonymous
Feb 05, 2026
Rating: 3/5
Runs hot under load
Performance is good but the board gets warm during extended inference tasks. Recommend a heatsink.
Q: Does this support standard Python libraries?
A: It supports CanMV Python which includes many standard ML libraries optimized for the K230.
Q: Can I power this via GPIO?
A: We recommend using the Type-C port for stable power. GPIO power input is not advised for full load.
Q: What camera module is included?
A: This SKU includes a standard OV5640 camera module compatible with the MIPI interface.
Q: Does it work with OpenMV IDE?
A: It uses the CanMV IDE which is similar but specifically optimized for the K230 chipset.
Q: Is there a heatsink included?
A: No, the board does not come with a heatsink. We recommend purchasing one separately for heavy AI workloads.