Gemini 3 Flash Preview

Gemini 3 Flash Preview is a new model in Google's gemini-flash line. It brings frontier-style multimodal reasoning to cheaper runs, according to the maker. It is a preview release dated 2025-12-17, succeeding Gemini 2.5 Flash. It has a context window of 1,048,576 tokens and a maximum output of 65,536 tokens. Input pricing starts at $0.07 per million tokens, output at $0.43 per million, and cached input at $0.025 per million. The weights are not open. It accepts audio, image, pdf, text, and video, and returns text. It supports reasoning mode and tool calling.

Against Gemini 2.5 Flash: input $0.09 to $0.07, output $0.71 to $0.43, cached input $0.021 to $0.025.

Range across the hosts that serve it: up to $0.70 per 1M input tokens.

• ·1M token context window for long inputs
• ·64K token max output for detailed answers
• ·Accepts audio, image, pdf, text, video
• ·Reasoning mode and tool calling enabled
• ·Input from $0.07 per million tokens

• Reach for it for multimodal reasoning on large inputs
• Reach for it for processing long documents and videos
• Reach for it for cost-sensitive high-volume inference
• Reach for it for tool-augmented agent workflows

29 hosts serve this model at the price above · the maker's documentation

The registry keeps the six most recent versions of a line. Older ones stay in the file and can be surfaced without a rebuild.

What is the context window?

1,048,576 tokens, same as the previous Gemini 2.5 Flash.

Does it support image input?

Yes, it accepts audio, image, pdf, text, and video, and returns text.

What is the price?

From $0.07 per million input tokens and $0.43 per million output tokens. Cached input is $0.025 per million.