All models

Gemma 4 31B

Google's downloadable model, small enough to run yourself. It understands pictures and video. A free listing exists on OpenRouter.

Last updated October 16, 2026.

Costs here use the prices AI companies charge software developers, so they show relative cost, not your monthly bill. See what ChatGPT, Claude and Gemini charge you.

Made by Google, the company behind Gemini. Free to download.

What it costs

1,000-word essay: under 1 cent
100 chat messages: $0.25 (about 25 cents)
50-page PDF summary: $0.01 (about 1 cent)

Based on list prices and the assumptions on How we work. Your real cost will differ.

At a glance

Smartness: ★☆☆☆☆ (1 of 5) (our banding of the Artificial Analysis score, where available)
Reads at once: about 390 pages (rule of thumb)
Understands: pictures, text, video
Free to download and run yourself: Yes
Free to use: A free developer version exists, with limits. Free app plans

App plans and prices for Gemini: see plans.

Best for

running it yourselfsmall projects

Watch out for

Well below the top models in quality.

Appears in

Best AI models you can download for free

Frequently asked

How much does Gemma 4 31B cost to use?
A 1,000-word essay costs about under 1 cent; summarizing a 50-page PDF about $0.01 (about 1 cent). These are developer (API) prices, not an app subscription.

What is Gemma 4 31B best for?
Google's downloadable model, small enough to run yourself. It understands pictures and video. A free listing exists on OpenRouter.

How much can Gemma 4 31B read at once?
About 390 pages (a rule of thumb: 262,144 tokens at roughly 500 words a page).

Can I download Gemma 4 31B for free?
Yes, the model files are published, though running a large one needs powerful hardware.

Under the hood

Maker	Google
Price per 1M input tokens	$0.20
Price per 1M output tokens	$0.40
Context size	262K tokens (262,144)
Artificial Analysis Intelligence Index	14.7 (source, via OpenRouter)
Open weights	Yes (Hugging Face: google/gemma-4-31B-it)
Speed	Not measured yet
Listed	2026-04-02

Maker's description (technical)

Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output. Features a 256K token context window, configurable thinking/reasoning mode, native function...

View on OpenRouter. Our verdict above is editorial, not benchmarked by us.