Tech · World
AI summary: catch up in 20 seconds
Summarising the coverage…
Generated by AI from the 4 sources below · always verify with the originals
The full picture
AI-synthesised from 4 sources · no editorial bias
Synthesis generating…
21 July
The Frozen v2 chip significantly boosts Gemini's performance, reportedly by ten times.
AI summary: catch up in 20 seconds
Generated by AI from the 4 sources below · always verify with the originals
The full picture
AI-synthesised from 4 sources · no editorial bias
Google is developing a new AI server chip, codenamed Frozen v2, aimed at enhancing the efficiency of its Gemini models. The chip is designed to integrate the architecture of the Gemini models directly into the hardware, which is expected to increase performance significantly—reports suggest it could be six to ten times more efficient than current tensor processing units (TPUs).
The anticipated release of the Frozen v2 chip is scheduled for 2028, and it is part of Google's broader effort to reduce AI inference costs. By leveraging this new chip within its data centers, Google aims to improve the overall effectiveness of its artificial intelligence capabilities, enabling more efficient operation of its Gemini models.
Skip to content
Comments · 0 (subscribers)
Sign in to join the discussion.
No comments yet. Be the first to join the discussion.