by Google DeepMind

Gemini 3.7 Flash.

multimodal closed 1M ctx
Cheapest input
$0.75/M
on DeepInfra
Cheapest output
$3.75/M
on DeepInfra
Hosted equiv.
~$1.35/hr
@ 100 tok/s on DeepInfra
Description

About Gemini 3.7 Flash.

Gemini 3.7 Flash is a deprecated model in Google's Gemini 3 series of multimodal reasoning models, replaced by Gemini 3.8 Flash. It can no longer be selected in Google AI Studio, and API requests for it are routed to 3.8 Flash. It took text, images, video, audio and PDF, returned text, and had a 1,048,576-token input limit. Thinking levels were low, medium and high; minimal returned an error.

Context window

How much it can remember.

1M tokens ≈ 786,432 English words
4K 32K 128K 1M
Capabilities

What it can do.

· Vision input
· Audio input
· Video input
· Function calling
· Tool use
· JSON mode
✓ Streaming
· Fine-tuning