--- title: "Introducing Gemini 3.7 Flash" url: https://daily.dev/posts/introducing-gemini-3-7-flash-pomclri8w source_url: https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-3-7-flash type: article source: "DeepMind" published: 2026-08-13T17:26:01.811Z updated: 2026-08-14T00:33:36.437Z tags: ["google", "llm", "ai-agents", "google-gemini"] reading_time: 4 upvotes: 1 comments: 0 language: 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. # Introducing Gemini 3.7 Flash **[DeepMind](https://daily.dev/sources/dm)** · 4 min read · 1 upvotes · 0 comments ## Summary Google introduces Gemini 3.7 Flash, arriving three weeks after 3.6 Flash, positioned as its most capable workhorse model yet for coding and agentic workflows. It shows notable gains over 3.6 Flash on FrontierCode (43.6% vs 34.4%), DeepSWE v1.1 (65.3% vs 49.0%), WebDev Arena Elo (1588 vs 1538), GDP.pdf (34.0% vs 22.0%), and AutomationBench (30.4% vs 17.0%). It launches at an introductory price of $0.75/1M input and $3.75/1M output tokens, roughly half the prior Flash cost, available through end of year. Gemini Spark, the personal agent for AI Pro/Ultra subscribers in over 160 countries, is being upgraded to use 3.7 Flash starting immediately. The model also ships updated safety safeguards for CBRN and cyber misuse domains. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What is Gemini 3.7 Flash and how does it compare to Gemini 3.6 Flash for coding? Gemini 3.7 Flash is Google's updated workhorse model for coding and agents, released three weeks after 3.6 Flash. It scores 43.6% versus 34.4% on FrontierCode 1.1 Main and 65.3% versus 49.0% on DeepSWE v1.1, showing stronger debugging, issue resolution, and first-pass code accuracy than its predecessor. _Developers comparing model versions before switching can track releases like this one on daily.dev._ ### How much does the Gemini 3.7 Flash API cost per token? Gemini 3.7 Flash launches at an introductory price of $0.75 per 1 million input tokens and $3.75 per 1 million output tokens, roughly half the cost of the prior 3.6 Flash pricing. This introductory rate is available through the end of the year. _Teams budgeting for LLM API costs can follow pricing shifts like this on daily.dev._ ### How does Gemini 3.7 Flash perform on web development and UI generation tasks? Gemini 3.7 Flash generates more functional layouts and feature-complete apps in fewer prompts than 3.6 Flash, with strong design adherence when given a reference screenshot, image, or design system. It scores an Elo of 1588 versus 1538 for 3.6 Flash on Arena.ai's WebDev Arena leaderboard. _Developers picking a model for UI generation workflows can weigh benchmarks like these on daily.dev._ ## Community take How the wider developer community reacted, aggregated from 2 discussions and 240 comments across hackernews (as of 2026-08-14). **TL;DR:** 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. **Sentiment:** 25% positive · 45% mixed · 30% skeptical **The case for** - Praised as very fast, with end-to-end latency seen as its biggest real advantage over competitors. - Strong multimodal and video/audio understanding capabilities, including unique YouTube link ingestion, are highlighted as differentiators. - Some found it a solid, cheap workhorse for tasks like PDF/image data extraction and OCR. - A few use it effectively in a multi-model pipeline (e.g., translating verbose Opus output into readable text). **The pushback** - 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. - Getting a Google API key or navigating GCP/Vertex is widely described as unnecessarily complex and friction-heavy compared to OpenAI/Anthropic. - Skepticism that Google can't deliver a competitive frontier 'Pro' model and is settling for fast-but-not-best positioning. - Criticism of the 'introductory price' framing and pricing tables as a face-saving gesture after past price hikes damaged trust. - Concerns about lack of real-time billing/budget caps in Google's console leading to bill-shock risk. **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. **Hottest debate:** 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. - Whether Google's Flash line can keep pace with rapidly improving cheaper competitors like Luna and Sol. - How much the choice of harness/agent (e.g., opencode vs cursor) affects comparative benchmark results between models. **Highlights** > Other thoughts: I really think Google has fallen behind here. Even as a high speed offering (this build took ~7min, which is pretty good!), it wont be able to claim dominance for long with cerebras announcing the Sol preview today: https://www.cerebras.ai/blog/accelerating-gpt-5-6-sol-ultraf... . It's not a bad model by any means, but I just don't know what situation I'd reach for 3.7 Flash first for. Google really needs a differentiator, especially given how hard it is to get an API key from them. They can't be high friction and non-pareto. > — [jjcm on hackernews · 5 comments](https://news.ycombinator.com/item?id=49289937) > It's funny that they don't mention this at all in the marketing or tech specs when it's obviously the biggest selling point by far. Without this it would be completely irrelevant. Worth noting that OpenAI just announced that they got the full GPT 5.6 Sol model running on Cerebras at 750 tokens per second. No announcement of the pricing though... > — [modeless on hackernews · 4 comments](https://news.ycombinator.com/item?id=49289869) > The multimodal abilities are great, but if you deal with text only, what is the benefit of using this over DS V4 Flash/Pro? 13-26x cheaper with comparable intelligence, and available across many different inference providers. I fail to see the usecase where DS V4 Pro is not enough, but Flash 3.7 is - except multimodal. Luna is similar, and also 8x cheaper. Source: artificialanalysis The only benefit I can see is the speed, that looks to be outstanding, probably thanks to their TPUs. > — [euazOn on hackernews · 7 comments](https://news.ycombinator.com/item?id=49289194) > Ever since the insane discount with GPT-5.6 Luna, not much excites me anymore. I mean just look at the benchmarks, even though Gemini 3.7 Flash performs well on the DeepSWE 1.1, Luna (Max) still performs way better. I personally have stuck to Luna (Xhigh) because its been more than enough and does not bloat up the context window too fast with reasoning tokens. https://deepswe.datacurve.ai > Starting January 1, 2027, $1.50/1M input tokens and $7.50/1M output tokens will apply. Compare this to Luna which is at $0.2/1M input ($0.02 cached) and $1.2/1M output. https://developers.openai.com/api/docs/models/gpt-5.6-luna > — [Alifatisk on hackernews · 6 comments](https://news.ycombinator.com/item?id=49289829) > You need to create a Google Cloud project to create an api key and when you try to create one you very often get error messages like: “Failed to create project, The request is suspicious. Please try again” or “ You do not have permission to create a key in this project”. You can then navigate multiple screens in GCP to make it work but it’s a hassle compared to any other provider (OAI/Ant/OpenRouter or any of the Chinese labs). > — [1bm on hackernews · 1 comments](https://news.ycombinator.com/item?id=49290946) > > 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) **Source threads** - [hackernews](https://news.ycombinator.com/item?id=49288847) · 46 points · 14 comments - [hackernews](https://news.ycombinator.com/item?id=49289112) · 314 points · 226 comments --- Tags: [#google](https://daily.dev/tags/google), [#llm](https://daily.dev/tags/llm), [#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/introducing-gemini-3-7-flash-pomclri8w)