I've been trying to find some benchmarking data for the AKIDA chip. I understand that BrainChip has had samples available for several months. So I find it strange that don't seem to be any hard numbers for performance. I don't mean accuracy, but rather how long the chip takes to process information, and power utilization. In principle, you can write code using the ADE and run it on the chip. If so, it should be straightforward to compare the chip with other solutions including NVIDIA Jetson chips on a wide range of standard neural network applications.
Has anyone seen such comparisons?
-
- There are more pages in this discussion • 82 more messages in this thread...
You’re viewing a single post only. To view the entire thread just sign in or Join Now (FREE)
Add to My Watchlist
What is My Watchlist?
 (20min delay)
|
|||||
|
Last
21.8¢ |
Change
0.003(1.16%) |
Mkt cap ! $435.4M | |||
| Open | High | Low | Value | Volume |
| 21.5¢ | 22.5¢ | 21.0¢ | $1.939M | 8.850M |
Buyers (Bids)
| No. | Vol. | Price($) |
|---|---|---|
| 34 | 1563038 | 21.5¢ |
Sellers (Offers)
| Price($) | Vol. | No. |
|---|---|---|
| 22.0¢ | 1082351 | 19 |
View Market Depth
| Last trade - 13.42pm 18/07/2025 (20 minute delay) ? |
| BRN (ASX) Chart |