Google Gemini 3.7 Flash
Google's newest Flash workhorse, a more capable successor to 3.6 Flash with stronger software engineering, better document comprehension and more disciplined tool use, at half the price per token.
Google's Gemini 3.7 Flash is a cloud text model. It is multimodal: alongside a text prompt it accepts images, then replies with generated text. Its context window handles up to 1.05M input tokens and up to 66K output tokens. Generation can be tuned with system instructions. It runs through Runware using your own API key, from $0.75 per million input tokens.
- Released
- 1 month ago (Aug 2026)
- Pricing
- $0.75 / 1M in · $3.75 / 1M out
- Inputs
- Text, Images
- Context window
- 1.05M in · 66K out
- Controls
- System prompt
Examples
Generated with Gemini 3.7 Flash via Runware. The same three prompts run against every text model in the catalog, shown verbatim — so the only thing that changes between two models’ answers is the model.
Explain what an API is to someone who has never written code. Do not use a restaurant or waiter analogy. Plain prose only — no headings, bullet points, or markdown formatting. Maximum 50 words.
An API is a secure bridge that allows two different computer programs to talk to each other and share information. For example, when you check the weather on your phone, an API requests the latest forecast from a remote weather system and immediately displays it on your screen.
A two-person startup is deciding whether to build their own user authentication or use a hosted provider. Give them a clear recommendation, then the single strongest argument against your own recommendation. Plain prose only — no headings, bullet points, or markdown formatting. Maximum 150 words.
You should use a hosted authentication provider. As a two-person team, your most critical asset is time, and your absolute priority must be finding product-market fit. Building custom authentication is deceptively complex, requiring you to implement and maintain password hashing, account recovery, session management, token validation, and regulatory compliance. Offloading these responsibilities lets you launch weeks faster while ensuring your users are protected by battle-tested infrastructure. The single strongest argument against this recommendation is the risk of painful vendor lock-in and steep pricing cliffs. Hosted providers can become exorbitantly expensive as your monthly active user count grows, and migrating sensitive user credentials, active sessions, and identity metadata away from a proprietary platform to an in-house solution later is a complex, high-risk engineering undertaking.
Write a briefing for a non-technical executive explaining why their company's spending on AI is hard to predict, and what they can do about it. Cover at least three distinct causes. Plain prose only — no headings, bullet points, or markdown formatting. Write exactly 205 words.
Artificial intelligence expenditures are notoriously difficult to forecast because costs behave fundamentally differently from traditional, fixed enterprise software licenses. First, most commercial platforms utilize dynamic, consumption pricing models where charges fluctuate unpredictably based on daily query volumes and the variable length of prompt interactions. Second, internal engineering teams constantly experiment with different foundational models, data pipelines, and custom training processes, triggering massive, unbudgeted compute surges during development phases. Third, the relentless pace of vendor updates and shifting provider fees creates persistent price instability across core external integrations. To effectively mitigate this financial unpredictability, executive leadership must establish disciplined operational governance across all business units. The organization should begin by implementing automated spending alerts, strict daily budget caps, and continuous monitoring to proactively prevent unexpected, costly cloud billing spikes. Furthermore, senior leaders must enforce granular cost attribution, making individual department managers directly accountable for the computational resources their respective teams consume. Finally, establishing a centralized review committee will ensure every proposed initiative is rigorously measured against clear return on investment benchmarks before deploying to live production. By uniting technical guardrails with departmental accountability, the company can successfully transform erratic artificial intelligence expenditures into a stable, strategic investment that advances corporate goals without compromising baseline fiscal health.
Google and Google DeepMind build the Gemini family of multimodal models, the Imagen and Nano Banana image models, the Lyria music models, and the Veo video models.
deepmind.google ↗