Google DeepMind 发布新的 Gemini Flash 模型，以提高效率和节省成本

报道来源 [13]

1. X — Google DeepMind TIER_1 English(EN) · GoogleDeepMind · 2026-07-21 15:29

   Gemini 3.6 Flash and 3.5 Flash-Lite are rolling out now in the @GeminiApp.

   Gemini 3.6 Flash and 3.5 Flash-Lite are rolling out now in the @GeminiApp. Developers can start building via the API in @GoogleAIStudio and @AndroidStudio. Gemini 3.5 Flash Cyber will be exclusively available via CodeMender soon as part of a limited-access pilot program. Find …

2. Google DeepMind TIER_1 English(EN) · 2026-07-21 15:16

   Introducing Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber

   We’re introducing new Gemini models, including Gemini 3.6 Flash, 3.5 Flash-Lite and 3.5 Flash Cyber.

3. Google DeepMind TIER_1 English(EN) · 2026-07-17 15:00

   Introducing Gemini 3.5 Flash Cyber

   Google introduces Gemini 3.5 Flash Cyber, a lightweight cybersecurity model to find and patch vulnerabilities.

4. Ars Technica — AI TIER_1 English(EN) · Ryan Whitwam · 2026-07-21 16:58

   Google announces Gemini 3.6 Flash and cybersecurity AI, teases 3.5 Pro and Gemini 4

   There are new 3.6 and 3.5 models today, but Google is already training Gemini 4.

5. Hacker News — AI stories ≥50 points TIER_1 English(EN) · logickkk1 · 2026-07-21 15:17

   Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber

6. MarkTechPost TIER_1 English(EN) · Asif Razzaq · 2026-07-21 17:45

   Google Releases Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber: A Cheaper, More Token-Efficient Flash Tier Built for Agentic Workloads

   <p>Google released Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber on July 21, 2026. The Flash tier gets cheaper and more token-efficient, with 3.6 Flash cutting output tokens 17% and dropping its output price to $7.50 per 1M. Flash-Lite runs at 350 tokens/sec, while gated …

7. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] · 2026-07-21 19:10

   Google releases Gemini 3.6 Flash Cyber, a cost-efficient AI model for security vulnerability detection and patching, positioned as an alternative to larger ente

   Google releases Gemini 3.6 Flash Cyber, a cost-efficient AI model for security vulnerability detection and patching, positioned as an alternative to larger enterprise models. # AI # Automation Source: The Verge AI https://www. theverge.com/tech/968572/googl e-gemini-flash-cyber-a…

   链接 theverge.com/…/google-gemini-flash-cyber-…
8. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] · 2026-07-21 18:50

   Google introduces three new Gemini models: 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber. Specialized options for performance and cybersecurity. # AI # Automat

   Google introduces three new Gemini models: 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber. Specialized options for performance and cybersecurity. # AI # Automation Source: Google AI Blog https:// blog.google/innovation-and-ai/ models-and-research/gemini-models/gemini-3-6-flash-3-…

   链接 blog.google/…/gemini-3-6-flash-3-5-flash-… blog.google/innovation-and-ai
9. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] · 2026-07-21 18:30

   Google DeepMind expands its AI model line with three new Gemini variants: 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber. Efficiency and specialization for ente

   Google DeepMind expands its AI model line with three new Gemini variants: 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber. Efficiency and specialization for enterprise workflows appear to be the focus. # AI # Automation Source: Google DeepMind Blog https:// deepmind.google/blog/in…

   链接 blog.google/…/gemini-3-6-flash-3-5-flash-…
10. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] · 2026-07-21 16:09

    Google has released Gemini 3.6 Flash, 3.5 Flash-Lite, and a restricted 3.5 Flash Cyber model, primarily targeting the cost and latency problems that come with r

    Google has released Gemini 3.6 Flash, 3.5 Flash-Lite, and a restricted 3.5 Flash Cyber model, primarily targeting the cost and latency problems that come with running AI agents continuously. https://www. artificialintelligence-news.co m/news/googles-gemini-3-6-flash-targets-enter…

11. Mastodon — mastodon.social TIER_1 Polski(PL) · [email protected] · 2026-07-21 17:20

    Google releases Gemini 3.6 Flash and 3.5 Flash-Lite. Faster AI for less money, but only a few will get the most powerful tool Google presented

    Google udostępnia Gemini 3.6 Flash i 3.5 Flash-Lite. Szybsze AI za mniejsze pieniądze, ale najmocniejsze narzędzie dostaną tylko nieliczni Google zaprezentowało trzy nowe warianty swoich modeli sztucznej inteligencji: Gemini 3.6 Flash, Gemini 3.5 Flash-Lite oraz specjalistyczny G…

    链接 blog.google/…/gemini-3-6-flash-3-5-flash-…
12. Mastodon — mastodon.social TIER_1 Polski(PL) · aisight · 2026-07-21 16:16

    Google introduces new Gemini 3.6 Flash and 3.5 Flash-Lite models, focusing on radical cost cutting and token efficiency. With prices starting from 0,

    Google wprowadza nowe modele Gemini 3.6 Flash i 3.5 Flash-Lite, stawiając na radykalne cięcie kosztów i oszczędność tokenów. Dzięki cenom zaczynającym się od 0,30 USD za milion tokenów, gigant chce zdominować rynek autonomicznych agentów. # si # ai # sztucznainteligencja # wiadom…

    链接 aisight.pl/…/google-optymalizuje-koszty-a… aisight.pl/…/google-opal-aplikacja-kodowa…
13. Mastodon — mastodon.social TIER_1 English(EN) · ngate · 2026-07-21 15:22

    Introducing the latest in # AI redundancy: Gemini 3.6 # Flash and its two slightly less robust siblings, because who doesn't need three versions of the same thi

    Introducing the latest in # AI redundancy: Gemini 3.6 # Flash and its two slightly less robust siblings, because who doesn't need three versions of the same thing? 🤖✨ Prepare to be dazzled by the # groundbreaking # innovation of slightly different numbers. 🎉 https:// blog.google/…

    链接 blog.google/…/gemini-3-6-flash-3-5-flash-… blog.google/innovation-and-ai