Gemini 3.7 Flash reaches GA for coding and agent workflows

Google has made Gemini 3.7 Flash (gemini-3.7-flash) generally available. The August 13, 2026 release is positioned as Google’s most capable Flash workhorse model yet for coding and agents.

What is new in 3.7 Flash

Google says Gemini 3.7 Flash improves software engineering, web development, knowledge work, and agentic workflows. Its developer guide lists support for text, image, video, audio, and PDF inputs; a 1,048,576-token input limit; a 65,536-token output limit; function calling; code execution; search and Maps grounding; URL context; structured outputs; and configurable thinking levels of low, medium, and high.

The model page lists computer use as supported in preview. Live API and image generation are not supported in the current model specification.

Introductory pricing

Google’s launch pricing is $0.75 per million input tokens and $3.75 per million output tokens through December 31, 2026. Google’s latest-model guide says the same introductory rate is also applied to Gemini 3.6 Flash; after the promotion, the listed standard rates are $1.50 per million input tokens and $7.50 per million output tokens.

Availability across Google’s developer stack

Gemini 3.7 Flash is available through the Gemini API and Google AI Studio, and Google’s announcement also names Android Studio, Google Antigravity, Gemini Enterprise Agent Platform, and Gemini Enterprise. Google says Gemini Spark uses 3.7 Flash for Google AI Pro and Ultra subscribers in supported countries.

The latest-model guide also says 3.7 Flash is now the default model powering the Antigravity agent in Gemini Managed Agents and the Google Antigravity SDK.

Migration notes

Google recommends updating the target model string to gemini-3.7-flash. The migration guidance also calls out several Gemini 3.x requirements, including replacing thinking_budget with the thinking_level enum and removing deprecated sampling parameters such as temperature, top_p, and top_k.

If you are migrating an existing integration, review the full model guide rather than changing only the model ID: Google’s instructions also cover turn validation, function-call formatting, and SDK/thought-signature requirements.

Sources

• Google AI for Developers — “Release notes | Gemini API” (August 13, 2026): https://ai.google.dev/gemini-api/docs/changelog
• Google AI for Developers — “Gemini 3.7 Flash” model page: https://ai.google.dev/gemini-api/docs/models/gemini-3.7-flash
• Google AI for Developers — “What’s new in Gemini 3.7 Flash”: https://ai.google.dev/gemini-api/docs/latest-model
• Google — “Introducing Gemini 3.7 Flash” (August 13, 2026): https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-3-7-flash/
• Google for Developers on X — official announcement: https://x.com/googledevs/status/2087951018973962410