DataLearner logo

Gemini 3.1 Pro PreviewvsGemini 3.0 Pro (Preview 11-2025)

Across 15 shared benchmarks, Gemini 3.1 Pro Preview leads overall: Gemini 3.1 Pro Preview wins 12, Gemini 3.0 Pro (Preview 11-2025) wins 3, with 0 ties and an average score difference of +7.84.

Google Deep Mind
Gemini 3.1 Pro Preview

Google Deep Mind · 2026-02-20 · Multimodal model

Google Deep Mind
Gemini 3.0 Pro (Preview 11-2025)

Google Deep Mind · 2025-11-18 · Multimodal model

Gemini 3.1 Pro Preview12 wins(80%)(20%)3 winsGemini 3.0 Pro (Preview 11-2025)

Benchmark scores

Grouped by capability, sorted by largest gap within each. 15 shared benchmarks.

General Knowledge

Gemini 3.1 Pro Preview 4/4
BenchmarkGemini 3.1 Pro PreviewGemini 3.0 Pro (Preview 11-2025)Diff
ARC-AGI-277.109 / 62Thinking High (No Tools)45.1026 / 62+32
LiveBench79.933 / 115Thinking High (No Tools)73.3924 / 115Thinking High (No Tools)+6.54
HLE51.4022 / 172Thinking High (With Tools)45.8040 / 172+5.60
GPQA Diamond94.303 / 187Thinking High (No Tools)93.805 / 187+0.50

Agent Level Benchmark

Gemini 3.1 Pro Preview 2/2
BenchmarkGemini 3.1 Pro PreviewGemini 3.0 Pro (Preview 11-2025)Diff
τ²-Bench90.802 / 43Thinking High (With Tools)85.408 / 43+5.40
τ²-Bench - Telecom99.301 / 35Thinking High (With Tools)985 / 35+1.30

AI Agent - Tool Usage

Gemini 3.1 Pro Preview 2/2
BenchmarkGemini 3.1 Pro PreviewGemini 3.0 Pro (Preview 11-2025)Diff
Terminal Bench 2.068.508 / 47Thinking High (With Tools)56.9025 / 47+11.60
MCP-Atlas78.209 / 27Thinking High (With Tools)70.3015 / 27Normal (With Tools)+7.90

Coding and Software Engineer

Even 2/2
BenchmarkGemini 3.1 Pro PreviewGemini 3.0 Pro (Preview 11-2025)Diff
SWE-bench Verified80.6011 / 112Thinking High (With Tools)76.2036 / 112+4.40
LiveCodeBench91.703 / 123Thinking High (With Tools)922 / 123-0.30

Math and Reasoning

Gemini 3.0 Pro (Preview 11-2025) 2/2
BenchmarkGemini 3.1 Pro PreviewGemini 3.0 Pro (Preview 11-2025)Diff
FrontierMath - Tier 416.7020 / 80Normal (No Tools)18.8016 / 80-2.10
FrontierMath36.9011 / 60Thinking High (No Tools)3810 / 60-1.10

AI Agent - Information Search

Gemini 3.1 Pro Preview 1/1
BenchmarkGemini 3.1 Pro PreviewGemini 3.0 Pro (Preview 11-2025)Diff
BrowseComp85.905 / 53Thinking High (With Tools + Internet)59.2038 / 53+26.70

Claw-style Agent Evaluation

Gemini 3.1 Pro Preview 1/1
BenchmarkGemini 3.1 Pro PreviewGemini 3.0 Pro (Preview 11-2025)Diff
Pinch Bench86.7010 / 37Thinking (With Tools)70.7031 / 37Thinking (With Tools)+16

Commonsense Reasoning

Gemini 3.1 Pro Preview 1/1
BenchmarkGemini 3.1 Pro PreviewGemini 3.0 Pro (Preview 11-2025)Diff
Simple Bench79.602 / 63Normal (No Tools)76.405 / 63Thinking (No Tools)+3.20

Specs

FieldGemini 3.1 Pro PreviewGemini 3.0 Pro (Preview 11-2025)
PublisherGoogle Deep MindGoogle Deep Mind
Release date2026-02-202025-11-18
Model typeMultimodal modelMultimodal model
ArchitectureDenseDense
ParametersNot availableNot available
Context length1M1000K
Max output64K64K

API pricing

Prices use DataLearner records when available; missing fields are not inferred.

ItemGemini 3.1 Pro PreviewGemini 3.0 Pro (Preview 11-2025)
Text input$2 / 1M tokens$2 / 1M tokens
Text output$12 / 1M tokens$12 / 1M tokens

Summary

  • Gemini 3.1 Pro Previewleads in:General Knowledge (4/4), Agent Level Benchmark (2/2), AI Agent - Tool Usage (2/2), AI Agent - Information Search (1/1), Claw-style Agent Evaluation (1/1), Commonsense Reasoning (1/1)
  • Gemini 3.0 Pro (Preview 11-2025)leads in:Math and Reasoning (2/2)
  • Tied in:Coding and Software Engineer

On average across the 15 shared benchmarks, Gemini 3.1 Pro Preview scores 7.84 higher.

Largest single-benchmark gap: ARC-AGI-2 — Gemini 3.1 Pro Preview 77.10 vs Gemini 3.0 Pro (Preview 11-2025) 45.10 (+32).

Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.

Gemini 3.1 Pro Preview detailsGemini 3.0 Pro (Preview 11-2025) details·Customize in compare tool