[{"data":1,"prerenderedAt":30},["ShallowReactive",2],{"nr-en-google-gemini-3-7-flash-preise-senken":3},{"slug":4,"title":5,"dek":6,"date":7,"time":8,"publishedAt":9,"updated":10,"updatedAt":10,"dateFmt":11,"updatedFmt":10,"kind":12,"tier":13,"author":14,"authorName":15,"topics":16,"tracker":22,"trackerLabel":23,"headlineStat":24,"image":25,"ogImage":26,"imageAlt":5,"csv":10,"minutes":27,"words":28,"html":29},"google-gemini-3-7-flash-preise-senken","Google cuts Gemini 3.7 Flash by half – enterprise AI gets cheaper, Pro models slow down","Google launched Gemini 3.7 Flash with dramatic price cuts for enterprise deployments. Meanwhile, development on more powerful Pro models is slowing – a pattern now visible across the entire AI industry.","2026-08-14","12:49","2026-08-14T12:49:00+02:00","","August 14, 2026","news","standard","ideal-syka","Ideal Syka",[17,18,19,20,21],"AI pricing","Enterprise AI","Language models","Google Gemini","Market consolidation","\u002Fki-preis","AI pricing & tariffs","50% price reduction","\u002Fnewsroom\u002Fimg\u002Fgoogle-gemini-3-7-flash-preise-senken.webp","\u002Fog-nr\u002Fgoogle-gemini-3-7-flash-preise-senken.en.png",3,618,"\u003Cp>Google has launched Gemini 3.7 Flash and cut prices for enterprise deployments by roughly 50 percent. The model now costs \u003Cstrong>$0.75 per million input tokens and $3.75 per million output tokens\u003C\u002Fstrong> – significantly less than its predecessor Gemini 3.6 Flash, which arrived just three weeks earlier. The price reduction signals that Google feels competitive pressure in the enterprise segment and is making production environments more economically viable.\u003C\u002Fp>\n\u003Ch2>The essentials\u003C\u002Fh2>\n\u003Cul>\n\u003Cli>\u003Cstrong>New pricing\u003C\u002Fstrong>: Gemini 3.7 Flash costs about half of 3.6 Flash – Google responds to competitive pressure\u003C\u002Fli>\n\u003Cli>\u003Cstrong>Focus\u003C\u002Fstrong>: Coding, automation, and agent workflows; Google calls it its "most intelligent workhorse model yet for coding and agents"\u003C\u002Fli>\n\u003Cli>\u003Cstrong>Benchmark jumps\u003C\u002Fstrong>: FrontierCode score from 34.4% to \u003Cstrong>43.6%\u003C\u002Fstrong>, AutomationBench from 17.0% to \u003Cstrong>30.4%\u003C\u002Fstrong>\u003C\u002Fli>\n\u003Cli>\u003Cstrong>Diverging cadence\u003C\u002Fstrong>: Flash models receive frequent updates, Pro models follow slower development – next Pro release unclear\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch2>Coding and automation take center stage\u003C\u002Fh2>\n\u003Cp>Google positions Gemini 3.7 Flash as a tool for software development and multi-step workflows. According to the company, the model improves debugging, issue resolution, and first-pass code generation. Company benchmarks show significant jumps:\u003C\u002Fp>\n\u003Cdiv class=\"tbl-scroll\">\u003Ctable>\n\u003Cthead>\n\u003Ctr>\n\u003Cth>Benchmark\u003C\u002Fth>\n\u003Cth>3.6 Flash\u003C\u002Fth>\n\u003Cth>3.7 Flash\u003C\u002Fth>\n\u003Cth>Improvement\u003C\u002Fth>\n\u003C\u002Ftr>\n\u003C\u002Fthead>\n\u003Ctbody>\u003Ctr>\n\u003Ctd>FrontierCode 1.1 Main\u003C\u002Ftd>\n\u003Ctd>34.4%\u003C\u002Ftd>\n\u003Ctd>43.6%\u003C\u002Ftd>\n\u003Ctd>+9.2 pp\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>DeepSWE v1.1\u003C\u002Ftd>\n\u003Ctd>49.0%\u003C\u002Ftd>\n\u003Ctd>65.3%\u003C\u002Ftd>\n\u003Ctd>+16.3 pp\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>AutomationBench\u003C\u002Ftd>\n\u003Ctd>17.0%\u003C\u002Ftd>\n\u003Ctd>30.4%\u003C\u002Ftd>\n\u003Ctd>+13.4 pp\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>WebDev Arena (Elo)\u003C\u002Ftd>\n\u003Ctd>1538\u003C\u002Ftd>\n\u003Ctd>1588\u003C\u002Ftd>\n\u003Ctd>+50 points\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Fdiv>\n\u003Cp>Google states that the model generates "more complete web applications with fewer prompts" and "adheres better to design inputs." In knowledge-intensive domains like finance, law, and biosciences, Gemini 3.7 Flash achieved \u003Cstrong>34.0% on the GDP.pdf benchmark, up from 22.0%\u003C\u002Fstrong>.\u003C\u002Fp>\n\u003Cblockquote>\n\u003Cp>"Gemini 3.7 Flash delivers a noticeably improved developer experience over 3.6 Flash. It better adapts to roadblocks, clarifies intent when needed, and follows instructions with greater fidelity." — Google (company statement)\u003C\u002Fp>\n\u003C\u002Fblockquote>\n\u003Ch2>The split update strategy\u003C\u002Fh2>\n\u003Cp>What makes this news particularly significant: Google and other vendors are now pursuing a \u003Cstrong>two-track development strategy\u003C\u002Fstrong>. Flash models, designed for fast and cost-effective tasks, receive frequent updates – Gemini 3.7 Flash followed just three weeks after 3.6 Flash. More powerful Pro models, built for complex reasoning, are being developed more slowly.\u003C\u002Fp>\n\u003Cp>Google has provided no timeline for its next Pro release. CEO Sundar Pichai dodged questions about a Pro update during the recent earnings call. A similar pattern appears at DeepSeek, which this week introduced its V4-Pro model as a higher-end offering alongside the V4-Flash variant.\u003C\u002Fp>\n\u003Ch2>Token efficiency over benchmark faith\u003C\u002Fh2>\n\u003Cp>Analysts caution against over-interpreting benchmark numbers. Sanchit Gogia of Greyhound Research notes: "3.7 Flash delivers substantial improvements across software engineering, knowledge work, and web development workflows. These remain vendor benchmark claims until the new model accumulates sufficient independent production evidence."\u003C\u002Fp>\n\u003Cp>For enterprises, what matters is whether improvements translate to real operations. Amit Chandak, Chief Analytics Officer at Kanerika, emphasizes: "For production teams, the more relevant number is token efficiency. Reductions in token usage lower both latency and cost at scale." In other words: benchmark points don't count – what matters is fewer correction loops, less human oversight, and more reliable multi-step execution.\u003C\u002Fp>\n\u003Ch2>What this means for enterprises\u003C\u002Fh2>\n\u003Cp>The price cut makes AI applications more economical in the enterprise segment – especially for automation and coding tasks. At the same time, slower Pro development shows the market consolidating: vendors focus on what generates revenue (fast, cheap models) while research into genuine "advanced reasoning" loses momentum. For enterprises: now is a good moment to pilot Flash models – costs drop, maturity rises. But if you're waiting for fundamental AI breakthroughs, expect longer cycles.\u003C\u002Fp>\n\u003Ch2>Sources\u003C\u002Fh2>\n\u003Cul>\n\u003Cli>\u003Ca href=\"https:\u002F\u002Fwww.infoworld.com\u002Farticle\u002F4209622\u002Fgoogle-cuts-gemini-3-7-flash-prices-as-enterprise-ai-economics-diverge-and-pro-cadence-slows.html\">InfoWorld\u003C\u002Fa>\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Cp>\u003Cem>Editorially owned by \u003Ca href=\"\u002Fen\u002Fautor\u002Fideal-syka\">Ideal Syka\u003C\u002Fa>. Sources and method: \u003Ca href=\"\u002Fen\u002Fredaktion\">Newsroom & method\u003C\u002Fa>. Tips and corrections: \u003Ca href=\"mailto:ai@i6eal.de\">ai@i6eal.de\u003C\u002Fa>.\u003C\u002Fem>\u003C\u002Fp>\n",1786707792045]