BREAKING
Google's 'Frozen v2' Etches Gemini in Silicon
0x
tokens/watt low
0x
tokens/watt high
0
target year
How Freezing the Model Works
1Freeze architecture
2Etch weights in silicon
3Cut data movement
4Boost efficiency
TPU vs Frozen v2
TPUsince 2016
Programmable
Training and inference
Frozen v2~2028
Fixed-function ASIC
Inference only
The Bet and the Risk
Google Unconfirmed, Details Thin
AI NEWS BLITZ
Google is reportedly building a chip that hardwires its Gemini models straight into silicon.