

🚀 Power your edge AI with Google's lightning-fast TPU accelerator
The SOM Google Edge TPU ML Compute Accelerator delivers ultra-efficient machine learning inferencing at 4 TOPS using just 2 watts, compatible with standard Half-Mini PCIe slots. It supports major OS platforms including Debian, Ubuntu, and Windows 10 on both x86-64 and ARMv8 architectures, enabling seamless integration into legacy and modern systems. Optimized for TensorFlow Lite and AutoML Vision Edge, it empowers developers to deploy high-speed, custom ML models directly at the edge.
| ASIN | B0844WRL58 |
| Best Sellers Rank | 227,599 in Computers & Accessories ( See Top 100 in Computers & Accessories ) 959 in Network Cards |
| Guaranteed software updates until | unknown |
| Item model number | G650-04528-01 |
| Manufacturer | Google Coral |
| Product Dimensions | 5.08 x 5.08 x 2.54 cm; 9 g |
B**.
This works great on an old Intel NUC that I use to run Frigate NVR.
G**D
I eventually got it working but it cost me a small fortune buying several expensive ($80+) adaptors and in the end only one PCIe adaptor would work in a Lenovo machine I have. Nothing would work in any of the Dell computers I have. The drivers don't work with new Linux distributions. I expected so much better from Google. Google isn't what they used to be, only in it for the money now. Sad.
N**T
Working well for person detection / recognition.
H**G
Works, not very reliable. You will have issues with drivers.
D**N
This is a mini-PCIe card. Most hosts have regular PCIe slots, so you will need an adapter like this one works: https://www.amazon.com/dp/B0CDLWC3G6?psc=1&ref=ppx_yo2ov_dt_b_product_details
Trustpilot
5 days ago
3 days ago