ZMIME
Google · Released Apr 2, 2026

Gemma 4 31B IT

Largest Gemma 4 instruction model for open, self-hosted chat and reasoning.

  • Open weights
  • Reasoning
  • Tool calling
  • Vision
Capability
144.1ECI · #94 of 158
Input
$0.142per 1M tokens
Output
$0.40per 1M tokens
Context
262K33K max output
Pricing

What it costs

Input
$0.142per million tokens
Output
$0.40per million tokens
Blended (3:1)
$0.206cheaper than 75% of priced models

Typical monthly bills

Monthly usageEstimated cost
Side project2M input + 0.5M output tokens$0.484
Team assistant25M input + 5M output tokens$5.55
Production app250M input + 50M output tokens$55.50

Median of 42 paid API providers on models.dev; Google also offers it free on Google.

Capability

Independent benchmarks

144.1Epoch Capabilities Index
#94 of 158 scored models
68Median 147174

The shaded band is Epoch AI’s confidence range (141.7–146.6); the tick marks the median scored model.

  • GPQA DiamondGraduate-level science questions · minimal setting75.8%
  • OTIS Mock AIME 2024–2025Competition mathematics · minimal setting73.3%
  • SimpleQA VerifiedShort factual questions10.4%

Scores from Epoch AI, run independently of Google.

Specifications

The details

Lab
Google
Released
Apr 2, 2026
Knowledge cutoff
Jan 2025
Context window
262,144 tokens
Max output
32,768 tokens
Inputs
Text, Images
Output
Text
Reasoning
Can be switched on or off
Tool calling
Yes
Structured output
Yes
API model ID
gemma-4-31b-iton Google
Availability
52 API providerslisted on models.dev

Google model documentation

Same lab

More from Google

Questions

About Gemma 4 31B IT

How much does Gemma 4 31B IT cost?

Gemma 4 31B IT costs $0.142 input / $0.40 output per million tokens (median across 42 API providers; free on Google). At a 3:1 input-to-output mix that is $0.206 per million tokens, cheaper than 75% of the 469 priced models we track.

What is the context window of Gemma 4 31B IT?

Gemma 4 31B IT accepts up to 262,144 tokens per request and can write up to 32,768 tokens in one response.

How good is Gemma 4 31B IT?

Epoch AI gives Gemma 4 31B IT a Capabilities Index score of 144.1 (likely range 141.7–146.6), ranking it #94 of 158 models Epoch has scored. Epoch AI benchmark results: GPQA Diamond 75.8%, OTIS Mock AIME 2024–2025 73.3%, SimpleQA Verified 10.4%.

Is Gemma 4 31B IT open source?

Yes. Google publishes the weights, so it can be downloaded and self-hosted.

What inputs does Gemma 4 31B IT support?

Gemma 4 31B IT accepts text and images and replies in text. It is a reasoning model, supports tool calling and can return structured JSON output.

When was Gemma 4 31B IT released?

Google released Gemma 4 31B IT on Apr 2, 2026. Its training data runs to Jan 2025.

What are the best alternatives to Gemma 4 31B IT?

The closest current models from other labs on capability, price and release date are Qwen3.5 Flash (Alibaba (Qwen), ECI 143.9, $0.10 / $0.40), DeepSeek V4 Flash (DeepSeek, ECI 146.8, $0.14 / $0.28), GPT-5.4 nano (OpenAI, ECI 146.3, $0.20 / $1.25) and MiniMax-M3 (MiniMax, ECI 147.9, $0.30 / $1.20).