Google
Released February 19, 2026

Gemini 3.1 Pro Preview

Google Gemini 3.1 Pro Preview is a Google model for better thinking, reliable multi-step tool use, structured outputs, and software engineering.

Inputs
Audio
File
Image
Text
Video
Outputs
Text

Model Overview

Capabilities, design details, and architectural traits

Gemini 3.1 Pro Preview - software engineering and tool-use model

Gemini 3.1 Pro Preview is built to refine the Gemini 3 Pro series for harder tasks. It emphasizes better thinking, more grounded answers, and reliable multi-step execution.

Distinctive traitWhat is documented
Software engineering focusOptimized for software engineering behavior and usability.
Reliable multi-step workBuilt for agentic workflows that require precise tool use and reliable multi-step execution.
Grounded responsesDescribed as more grounded and factually consistent.
Thinking and toolsSupports thinking, code execution, file search, function calling, search grounding, structured outputs, URL context, and Google Maps grounding.

Model behavior

The model is presented as a preview step for improving the Gemini 3 Pro series. Its documented pattern is to make advanced reasoning practical in real-world domains rather than only producing a direct answer.

Developer shape

The docs also list support for text, image, video, audio, and PDF inputs, plus Batch API, Flex inference, and Priority inference. A separate custom-tools endpoint is documented for bash and custom tool workflows.

Benchmark Performance

Independent evaluations · Artificial Analysis

29.7%
Intelligence
68.8%
Coding Index
8.2%
Agentic Index

Accuracy & Capability Details

GPQA - Graduate Science94.1%
Humanity's Last Exam47.0%
SciCode - Scientific Coding58.7%
Instruction Following77.1%
Long Context Reasoning82.0%
τ²-Bench - Agentic Tasks95.6%
TerminalBench - System Control53.8%

Compare Models Side-by-Side

Evaluate specifications, pricing, and independent benchmark indices

Model Details
General Info
ProviderGoogleAnthropicOpenAI
Release DateFebruary 19, 2026November 9, 2026November 17, 2026
Knowledge Cutoff---
Context & Limits
Context Window1.0M1M
1.1M
Best Context Window
Pricing (per 1M tokens)
Input Pricing
$2
Best Input Pricing
$8$8
Output Pricing
$12
Best Output Pricing
$40$40
Modalities
Inputs
audiofileimagetextvideo
textimagefile
textimagefile
Outputs
text
text
text
Benchmarks (0-100)
Intelligence Index29.7
64.6
Best Intelligence Index
62.1
Coding Index68.8--
Agentic Index8.2--

Benchmark comparison not available

Select at least two models that have available benchmark evaluation data to render the performance comparison charts.

Independent evaluation data provided by Artificial Analysis. To view the latest benchmarks and full details, visit their official site.

Explore more from Google

Other models by Google

Top AI Models

Leading alternatives by intelligence score

View all