Google · Model release
Gemini 3.6 Flash: the efficient multimodal workhorse
Gemini 3.6 Flash focuses on speed, multimodal input, and lower output usage for agentic and high-volume work.
What changed
- Google paired the new Flash tier with a cheaper Flash-Lite option for volume.
- The family accepts text, image, video, audio, and document inputs.
- Google frames the release around token efficiency rather than a universal leaderboard win.
Best fit
- Multimodal pipelines
- Document processing
- Fast agent loops
What to be careful about
- A large context window does not remove the need for retrieval and structure.
- Search grounding and other tools can add separate costs.
Dataset snapshot: Gemini 3.6 Flash is also represented in the checked comparison dataset. Prices and benchmark figures carry their own field dates.