Google

Gemini 3.1 Pro Preview

by Googletext+image+file+audio+video->text6 endpoints
Neura Intelligence Index
72.6/ 100
Rank
#68
Confidence
high
Benchmarks
12(31% coverage)

Overview

Gemini 3.1 Pro Preview is Google’s frontier reasoning model, delivering enhanced software engineering performance, improved agentic reliability, and more efficient token usage across complex workflows. Building on the multimodal foundation...

Capabilities

Text generation
Image understanding
Audio processing
Video understanding
Tool use / Function calling
Structured output
Extended reasoning
Prompt caching
Long context
File input
Top-P sampling
Stop sequences
Deterministic seed

Modalities

Input
AudioFileImageTextVideo
Output
Text

Technical Specifications

Context Window
1.0M tokens
Max Output
65.5K tokens
Tokenizer
Gemini
Uptime (24h)
9991.0%
Country
🇺🇸US
Release Date
February 19, 20267mo

Supported Parameters

include_reasoningmax_tokensreasoningreasoning_effortresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_p

Benchmark Performance

general

BenchmarkScoreSource
MMMLU
92.6
verified
MRCR v2
26.3
verified

coding

BenchmarkScoreSource
SWE-bench Verified
80.6
verified
SciCode
59
verified
SWE-Bench Pro
54.2
verified

reasoning

BenchmarkScoreSource
GPQA Diamond
94.3
verified
ARC-AGI v2
77.1
verified
Humanity's Last Exam
51.4
verified

multimodal

BenchmarkScoreSource
MMMU-Pro
80.5
verified

agent

BenchmarkScoreSource
BrowseComp
85.9
verified
MCP Atlas
69.2
verified
APEX Agents
33.5
verified

Performance Over Time

Pricing

Per 1M tokens
Input$2.00
Output$12.00
Cache read$0.20
Cache write$0.38
Audio$2.00
Reasoning$12.00
Image$2.00
Added
February 19, 2026
Last synced 9/21/2026