Google 的 "Frozen v2" 芯片将通过 Gemini 提高效率 6-10 倍，预计 2028 年发布 · 跟踪 5 个来源

报道来源 [8]

1. The Decoder TIER_1 English(EN) · Matthias Bastian · 2026-07-20 18:08

   谷歌的“Frozen v2”芯片据称将Gemini架构直接集成到硅中以提高效率

   <p><img alt="" class="attachment-full size-full wp-post-image" height="768" src="https://the-decoder.com/wp-content/uploads/2026/07/google_gemini-2.png" style="height: auto; margin-bottom: 10px;" width="1376" /></p> <p> Google is developing "Frozen v2," a server chip that bakes t…

2. Tom's Hardware TIER_1 English(EN) · Luke James · 2026-07-21 10:40

   Google reportedly developing 'Frozen v2' chip with Gemini's architecture etched into the silicon — engineers project 6 to 10 times more tokens per watt than latest TPUs

   Google is developing a server chip, informally dubbed "Frozen v2," that would etch part of its Gemini model's architecture directly into the silicon.

3. TechCrunch AI TIER_1 English(EN) · Lucas Ropek · 2026-07-20 21:21

   Google 正在研发新款 AI 芯片，旨在提升 Gemini 效率

   Alphabet, Google's parent company, is reportedly working on a new chip designed to make its Gemini models run much more efficiently.

4. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] · 2026-07-21 01:30

   谷歌正在打造定制AI芯片，以提高其Gemini模型的效率。#AI #自动化

   Google is building a custom AI chip to improve the efficiency of its Gemini models. # AI # Automation Source: TechCrunch AI https:// techcrunch.com/2026/07/20/goog le-is-working-on-a-new-ai-chip-designed-to-make-gemini-more-efficient/

   链接 techcrunch.com/…/google-is-working-on-a-n…
5. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] · 2026-07-20 19:50

   谷歌的“Frozen v2”芯片将Gemini架构嵌入硅中，效率提升6-10倍，目标是到2028年降低推理成本。#人工智能 #自动化

   Google's 'Frozen v2' chip embeds Gemini's architecture in silicon for 6-10x efficiency gains, targeting lower inference costs by 2028. # AI # Automation Source: The Decoder AI https:// the-decoder.com/googles-frozen -v2-chip-reportedly-bakes-geminis-architecture-directly-into-sil…

   链接 the-decoder.com/googles-frozen-v2-chip-re…
6. Mastodon — mastodon.social TIER_1 Deutsch(DE) · aisyndicate · 2026-07-21 06:10

   谷歌的Frozen-v2芯片旨在使Gemini架构与硬件紧密匹配，据消息人士称，其效率是TPU的6到10倍。这降低了推理成本。

   Googles Frozen-v2-Chip soll Gemini-Architektur hardwarenah abbilden und laut Quellen 6- bis 10-fache Effizienz gegenüber TPUs bieten. Das senkt Inference-Kosten drastisch, verschiebt aber die Hardware-Abhängigkeit weiter nach unten. https:// the-decoder.de/googles-frozen- v2-chip…

   链接 the-decoder.de/googles-frozen-v2-chip-sol…
7. Mastodon — mastodon.social TIER_1 English(EN) · [email protected] · 2026-07-20 21:21

   Google 正在研发一款新的 AI 芯片，旨在提高 Gemini 的效率

   Google is working on a new AI chip designed to make Gemini more efficient https://techcrunch.com/2026/07/20/google-is-working-on-a-new-ai-chip-designed-to-make-gemini-more-efficient/ # AI # Hardware # Chips

   链接 techcrunch.com/…/google-is-working-on-a-n…
8. Mastodon — mastodon.social TIER_1 Polski(PL) · aisight · 2026-07-20 19:37

   Google 设计 Frozen v2 处理器，将 Gemini 模型架构直接集成到硬件中。据称，新芯片的效率比现有芯片高出 10 倍。

   Google projektuje procesor Frozen v2, który integruje architekturę modelu Gemini bezpośrednio w sprzęcie. Nowy układ ma być do 10 razy wydajniejszy od obecnych jednostek TPU. # si # ai # sztucznainteligencja # wiadomości # informacje # technologia https:// aisight.pl/hardware/fro…

   链接 aisight.pl/…/frozen-v2-google-architektur…