Google ships Gemini 3.7 Flash for coding just 3 weeks after 3.6, at half the price

Google DeepMind has shipped Gemini 3.7 Flash barely three weeks after 3.6 Flash, an unusually fast cadence that Sundar Pichai framed as deliberate: 'Our Flash models are workhorses... we're shipping updates fast to get them in developers' hands.' The model targets coding and agentic workflows, delivering roughly 25% improved coding performance and gains across knowledge work and web development, and Google claims it outperforms rivals like Claude Sonnet 5 and GPT-5.6 Terra on coding benchmarks (developers cited FrontierCode 43.6% vs competitors' 34-41%).

The economics are the story. Gemini 3.7 Flash launches at 50% below 3.6 Flash's price through end of year, with rates rising to $1.50 per million input and $7.50 per million output tokens in 2027. AI Studio lead Logan Kilpatrick emphasized the model is available in the API, AI Studio, Antigravity, and Gemini chat for Pro and Ultra users, calling the intelligence increase in just three weeks a product of 'algorithmic improvements.'

The launch dominated Hacker News with 949 points and 482 comments — the day's biggest single launch discussion — with a consensus that it's 'better than 3.6 almost everywhere at half the launch price.' Google separately announced the Gemini app crossed one billion users, a distribution milestone unmatched by rivals.

The caveat, flagged by developers and Reuters alike: Google gave no timeline for its delayed flagship Gemini Pro model, drawing side-eye that Gemini leads on distribution and cheap workhorse models while rivals ship more powerful frontier releases. The one-billion-user milestone reads differently when the flagship keeps slipping. Watch whether the aggressive Flash cadence and pricing pressures competitors' margins, and whether Pro finally ships before year-end.