Google models

Google develops the Gemini models through Google DeepMind and offers them through the Gemini API, Google AI Studio and Google Cloud. As of October 2026 its stable Gemini text models are Gemini 3.8 Flash, Gemini 3.6 Flash and two Flash-Lite models, while Gemini 3.1 Pro is offered as a preview. Its open-weight Gemma 4 family, released under Apache 2.0, runs from on-device E2B and E4B models up to a 31B dense model and a 26B mixture-of-experts.

8 models trackedData checked October 10, 2026

Lineup

Model	Status	Weights	Tier	Context	Input / 1M	Output / 1M	Released
Gemini 3.8 FlashGemini 3	Current	Closed	Advanced	1M	$0.75	$3.75	Sep 2, 2026
Gemini 3.6 FlashGemini 3	Current	Closed	Advanced	1M	$0.75	$3.75	Jul 21, 2026
Gemini 3.5 Flash-LiteGemini 3	Current	Closed	Efficient	1M	$0.30	$2.50	Jul 21, 2026
Gemini 3.1 Flash-LiteGemini 3	Current	Closed	Efficient	1M	$0.25	$1.50	May 7, 2026
Gemma 4 31BGemma 4	Current	Open weights	Efficient	256K	$0.20	$0.40	Mar 31, 2026
Gemma 4 26B A4BGemma 4	Current	Open weights	Compact	256K	$0.07	$0.34	Mar 31, 2026
Gemma 4 E4BGemma 4	Current	Open weights	Compact	128K	—	—	Mar 31, 2026
Gemini 3.1 ProGemini 3	Preview	Closed	Frontier	1M	$2	$12	Feb 19, 2026

The lowest-priced current model is Gemma 4 26B A4B at $0.07 input and $0.34 output per million tokens. The most capable tier in the lineup is frontier, led by Gemini 3.1 Pro. The longest context window is 1M tokens (Gemini 3.8 Flash).

Frequently asked questions

What models does Google offer?

As of October 10, 2026, we track 8 Google models: Gemini 3.8 Flash, Gemini 3.6 Flash, Gemini 3.5 Flash-Lite, Gemini 3.1 Flash-Lite, Gemma 4 31B, Gemma 4 26B A4B, Gemma 4 E4B, Gemini 3.1 Pro. Status, prices and context windows for each are in the table on this page.

How do I access Google models?

Through Google's developer platform (ai.google.dev/gemini-api/docs), in the Gemini app, and for many models through cloud marketplaces and inference providers listed on each model page.

Does Google release open-weight models?

Yes: Gemma 4 under Apache 2.0 (earlier Gemma generations used the Gemma Terms of Use). Gemini models are proprietary.

Other providers

OpenAI Anthropic xAI Amazon (AWS)Meta Perplexity Mistral AI DeepSeek Alibaba (Qwen)Microsoft (Phi)NVIDIA Moonshot AI Z.ai (Zhipu AI)MiniMax IBM Ai2 (Allen Institute for AI)

Still choosing a model?

Send us your use case and constraints. We'll come back with a shortlist and an evaluation plan you can run on your own data.

Free and vendor-neutral: no model provider pays for placement.
Your details go only to partners you agree to.