--- title: Gemini 3.7 Flash leaked for days before Google finally... description: Gemini 3.7 Flash leaked repeatedly over about a week, spotted inside Gemini Enterprise, the Python GenAI SDK on GitHub, and the Google Cloud Console, before... canonical: https://daily.dev/posts/gemini-3-7-flash-leaked-for-days-before-google-finally-shipped-it-bpjkk2qve twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Gemini 3.7 Flash leaked for days before Google finally shipped it | daily.dev og:description: Gemini 3.7 Flash leaked repeatedly over about a week, spotted inside Gemini Enterprise, the Python GenAI SDK on GitHub, and the Google Cloud Console, before... og:url: https://daily.dev/posts/gemini-3-7-flash-leaked-for-days-before-google-finally-shipped-it-bpjkk2qve og:image: https://api.daily.dev/og/posts/BPjkK2Qve.png og:image:alt: Gemini 3.7 Flash leaked for days before Google finally shipped it og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Gemini 3.7 Flash leaked for days before Google finally shipped it **[Trends](https://daily.dev/sources/trends)** · 1 min read · 14 upvotes · 0 comments ## Summary Gemini 3.7 Flash leaked repeatedly over about a week, spotted inside Gemini Enterprise, the Python GenAI SDK on GitHub, and the Google Cloud Console, before Google officially confirmed general availability via Philipp Schmid. Aimed at coding and agent use cases, it shows benchmark jumps on DeepSWE (49.0% to 65.3%), FrontierCode (34.4% to 43.6%), and AutomationBench (17.0% to 30.4%). Pricing is $0.75 per million input tokens and $3.75 per million output, with a 50% discount through year end. It's now the default model across Managed Agents, Google AI Studio, Antigravity, and Gemini App Spark. ## Content Google dropped Gemini 3.7 Flash just three weeks after 3.6 Flash, and the timing is the story as much as the model itself. While the internet argues over benchmark deltas, the real subplot is Gemini 3.5 Pro, promised for June, still nowhere in sight, with Axios reporting DeepMind might skip it entirely and jump straight to Gemini 4. One piece put it bluntly: Google's cheap model is now two versions ahead of its own flagship. The numbers Google's leading with: DeepSWE v1.1 jumped from 49% to 65.3%, FrontierCode 1.1 from 34.4% to 43.6%, AutomationBench from 17% to 30.4%, Terminal-bench 3.0 from 5.4% to 14.9%. Pricing halved to $0.75/$3.75 per million input/output tokens, promotional until the end of 2026, then doubling. @VanquishTrader called it a pretty clean strategy read: Google is ## Questions this post answers ### What are the benchmark improvements for Gemini 3.7 Flash compared to the previous version? Gemini 3.7 Flash shows notable gains on agentic coding benchmarks: DeepSWE rose from 49.0% to 65.3%, FrontierCode went from 34.4% to 43.6%, and AutomationBench climbed from 17.0% to 30.4%. These are substantial jumps rather than marginal improvements, positioning the model to compete with OpenAI and Anthropic in agentic coding tasks. _Developers picking a coding-agent model can track how these benchmark claims hold up on daily.dev._ ### How much does Gemini 3.7 Flash cost per token? Gemini 3.7 Flash is priced at $0.75 per million input tokens and $3.75 per million output tokens, with a 50% discount applied through the end of the year. It has become the default model across Managed Agents, Google AI Studio, Antigravity, and Gemini App Spark, signaling a broad rollout rather than a niche release. _Teams comparing model costs for agentic workflows can follow pricing changes like this on daily.dev._ ## Community take How the wider developer community reacted, aggregated from 3 discussions and 257 comments across hackernews, x (as of 2026-08-19). **TL;DR:** Early reactions are enthusiastic about the model's quality and accuracy, though at least one person flags increased verbosity and hallucination compared to the prior version. Introducing Gemini 3.7 Flash: Reactions are mixed-to-lukewarm: the model's speed and multimodal/video capabilities are widely praised, but many see it as falling behind cheaper rivals like Luna, DeepSeek, and Sol on coding benchmarks, with continued frustration over Google's API/GCP onboarding friction and pricing tactics. Gemini 3.7 Flash: Discussion centers on Gemini 3.7 Flash's pricing quirk (a two-year . **Sentiment:** 26% positive · 45% mixed · 29% skeptical **The case for** - Seen as impressive and accurate, especially for nuanced/raw knowledge beyond just coding. - Praised as very fast, with end-to-end latency seen as its biggest real advantage over competitors. - Speed and end-to-end response time are widely praised as the model's key differentiator. - Perceived as less censored than competing models. - Strong multimodal and video/audio understanding capabilities, including unique YouTube link ingestion, are highlighted as differentiators. **The pushback** - Compared to the previous version, it reportedly hallucinates more and is much more verbose with the same instructions. - Many argue cheaper competitors like GPT-5.6 Luna, DeepSeek V4, and Sol offer comparable or better performance for coding/agentic tasks at a fraction of the cost. - The 2027 price-hike disclosure is widely mocked as pointless since the model will likely be obsolete by then. - Getting a Google API key or navigating GCP/Vertex is widely described as unnecessarily complex and friction-heavy compared to OpenAI/Anthropic. - Many feel Google mis-priced or overpriced 3.5 Flash previously and lost trust after a price hike, pushing users to competitors. **By community** - hackernews (mixed): Discussion splits between admiration for speed/multimodal strengths and frustration over pricing history, API friction, and being outpaced by cheaper competitors on coding benchmarks. - x (mixed): Reactions are largely impressed with the model's capability, but tempered by a concrete complaint about increased hallucination and verbosity versus the prior release. **Hottest debate:** Introducing Gemini 3.7 Flash: Whether Flash-tier speed and multimodal ability actually compensate for lagging behind cheaper rivals on coding/agentic benchmarks. **Open questions** - Why Google doesn't market its speed advantage more prominently despite it being called the biggest selling point. - Why does the introductory pricing expire so far in the future for a model likely to be obsolete within months? - Whether Google's Flash line can keep pace with rapidly improving cheaper competitors like Luna and Sol. - Is Google intentionally focusing on speed/cost over frontier intelligence, and if so, is that a viable long-term strategy? - How much the choice of harness/agent (e.g., opencode vs cursor) affects comparative benchmark results between models. **Highlights** > @_philschmid @ArtificialAnlys wow, that's impressive! not surprised, it's in line with my experience using it over the last few days. > — [intellectronica on x · 6 points](https://x.com/intellectronica/status/2090064904342437963) > @_philschmid @ArtificialAnlys Please keep it up. Less censored, insightful, very accurate for raw and nuanced knowledge, not only code, as the lobotomized competitors are today. Only thing is compared to 3.6, it hallucinates a lot more and is A LOT more verbose with same instructions > — [brunobevale on x · 4 points](https://x.com/brunobevale/status/2090069865805607117) > > 3.7 Flash is available through the end of the year at an introductory price of $0.75/1M input tokens and $3.75/1M output tokens. > Introductory pricing expires on December 31, 2026. Starting January 1, 2027, $1.50/1M input tokens and $7.50/1M output tokens will apply. > — [damsta on hackernews · 1 comments](https://news.ycombinator.com/item?id=49289238) > > introductory price They should call it 'face saving pricing after we realized just how terribly did we mis-price the flash 3.5' > since google betrayed us with those price hike, people already spent their time making their production pipeline less dependent on google since then. This is my first hand experience. I spent at least $3000 on gemini-3-flash-preview. And exactly $0 total on (3.5+3.6+3.7) > — [GodelNumbering on hackernews · 1 comments](https://news.ycombinator.com/item?id=49289350) > Another failed 3.5 pro run branded as 3.7 flash. It's getting sad. > — [AntonioEritas on hackernews · 1 comments](https://news.ycombinator.com/item?id=49289176) **Source threads** - [hackernews](https://news.ycombinator.com/item?id=49289112) · 314 points · 226 comments - [x](https://x.com/_philschmid/status/2090063976872751408) · 0 points · 17 comments - [hackernews](https://news.ycombinator.com/item?id=49288847) · 46 points · 14 comments --- Tags: [#ai](https://daily.dev/tags/ai), [#google](https://daily.dev/tags/google), [#ai-agents](https://daily.dev/tags/ai-agents), [#google-gemini](https://daily.dev/tags/google-gemini) [View this post on daily.dev](https://daily.dev/posts/gemini-3-7-flash-leaked-for-days-before-google-finally-shipped-it-bpjkk2qve) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Gemini 3.7 Flash leaked for days before Google finally shipped it","url":"https://daily.dev/posts/gemini-3-7-flash-leaked-for-days-before-google-finally-shipped-it-bpjkk2qve","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/gemini-3-7-flash-leaked-for-days-before-google-finally-shipped-it-bpjkk2qve"},"datePublished":"2026-08-13T17:05:39.098Z","dateModified":"2026-08-19T16:12:18.911Z","description":"Gemini 3.7 Flash leaked repeatedly over about a week, spotted inside Gemini Enterprise, the Python GenAI SDK on GitHub, and the Google Cloud Console, before...","image":"https://pbs.twimg.com/media/HPngj5OWUAA7lt6.jpg","thumbnailUrl":"https://pbs.twimg.com/media/HPngj5OWUAA7lt6.jpg","isAccessibleForFree":true,"articleSection":"Trends","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Trends","logo":"https://media.daily.dev/image/upload/s--ZfSp3asX--/f_auto,q_auto/v1780996004/logos/trends?_a=BAMAMiWQ0","url":"https://daily.dev/sources/trends"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/gemini-3-7-flash-leaked-for-days-before-google-finally-shipped-it-bpjkk2qve","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":14},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai,google,ai-agents,google-gemini","timeRequired":"PT1M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Trends","item":"https://daily.dev/sources/trends"},{"@type":"ListItem","position":3,"name":"Gemini 3.7 Flash leaked for days before Google finally shipped it"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/gemini-3-7-flash-leaked-for-days-before-google-finally-shipped-it-bpjkk2qve#faq","mainEntity":[{"@type":"Question","name":"What are the benchmark improvements for Gemini 3.7 Flash compared to the previous version?","acceptedAnswer":{"@type":"Answer","text":"Gemini 3.7 Flash shows notable gains on agentic coding benchmarks: DeepSWE rose from 49.0% to 65.3%, FrontierCode went from 34.4% to 43.6%, and AutomationBench climbed from 17.0% to 30.4%. These are substantial jumps rather than marginal improvements, positioning the model to compete with OpenAI and Anthropic in agentic coding tasks. Developers picking a coding-agent model can track how these benchmark claims hold up on daily.dev."}},{"@type":"Question","name":"How much does Gemini 3.7 Flash cost per token?","acceptedAnswer":{"@type":"Answer","text":"Gemini 3.7 Flash is priced at $0.75 per million input tokens and $3.75 per million output tokens, with a 50% discount applied through the end of the year. It has become the default model across Managed Agents, Google AI Studio, Antigravity, and Gemini App Spark, signaling a broad rollout rather than a niche release. Teams comparing model costs for agentic workflows can follow pricing changes like this on daily.dev."}}]} ```