Noometry

Model comparison

Gemini 2.5 Pro vs Gemini 3.1 Pro Preview

Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 45.0 on the Noometry Index.

Last verified . 52 shared benchmarks.

Gemini 2.5 Pro Google

45.0

Rank #75 Confirmed

Gemini 3.1 Pro Preview Google

56.7

Rank #23 Confirmed

Summary

  • They share 52 benchmarks with published results for both. Gemini 2.5 Pro scores higher in 2 categories and Gemini 3.1 Pro Preview in 8 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Gemini 3.1 Pro Preview leads 71.7 to 28.8.
  • The biggest single-benchmark swing is ARC-AGI-2: 4.9% for Gemini 2.5 Pro and 77.1% for Gemini 3.1 Pro Preview.
  • Gemini 2.5 Pro is cheaper at $1.25 / $10 per million input/output tokens, against $2 / $12 for Gemini 3.1 Pro Preview.

Side by side

Gemini 2.5 ProGemini 3.1 Pro Preview
ProviderGoogleGoogle
Noometry Index45.056.7
Released2025-03-252026-02-19
WeightsProprietaryProprietary
Context window1.05M1.05M
Max output66K66K
Input $ / M tokens$1.25$2
Output $ / M tokens$10$12
Results tracked7871

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding Too close to call

Gemini 2.5 Pro: 42.4 (#101), Gemini 3.1 Pro Preview: 42.5 (#99)

BenchmarkGemini 2.5 ProGemini 3.1 Pro Preview
SWE-bench Verified57.6%75.6%
LMArena WebDev12271447
SciCode42.8%58.9%
GSO3.9%22.6%
WeirdML54%72.1%
LMArena Coding14521484
ALE-Bench785.521,161
AlgoTune1.512.02
DeepSWE—11.7%
SWE-bench Verified (bash only)53.6%—
Aider Polyglot83.1%—
LiveBench Coding85.9%—
MirrorCode—8.9%
CadEval64%—

Agentic & Tool Use Gemini 3.1 Pro Preview leads

Gemini 2.5 Pro: 29.2 (#88), Gemini 3.1 Pro Preview: 37.7 (#34)

BenchmarkGemini 2.5 ProGemini 3.1 Pro Preview
Terminal-Bench32.6%80.2%
τ²-bench Banking13.7%26%
DeepResearch Bench42.8%47.8%
BALROG43.3%57%
LMArena Search11421211
METR Time Horizons55.4%77%
Vending-Bench 2573.643,774
APEX-Agents—35.3%
GDPval23.3%—
Remote Labor Index0.8%—
TheAgentCompany30.3%—
PostTrainBench—22%
ExploitBench—26.1%
GBAEval—0.8%
GDP.pdf—17%

Reasoning Gemini 3.1 Pro Preview leads

Gemini 2.5 Pro: 28.8 (#99), Gemini 3.1 Pro Preview: 71.7 (#12)

BenchmarkGemini 2.5 ProGemini 3.1 Pro Preview
ARC-AGI-24.9%77.1%
SimpleBench62.4%79.6%
ARC-AGI-141%98%
CritPt2%17.7%
Chess Puzzles20%55%
EnigmaEval5.6%36.8%
LMArena Hard Prompts14551485
DTBench82.4%97.1%
LMCA34.8%53.8%
Epoch Capabilities Index145.32154.77
ForecastBench61.359
Kagi LLM Benchmark70.3%—
NYT Connections (extended)—97.4%
Thematic Generalization—79.4%
EBR-Bench—14.3%
LiveBench Reasoning89.8%—
Mystery Game Puzzles—34%
LiveBench Data Analysis79.9%—
LiveBench82.3%—

Math Gemini 3.1 Pro Preview leads

Gemini 2.5 Pro: 32.5 (#213), Gemini 3.1 Pro Preview: 62.1 (#34)

BenchmarkGemini 2.5 ProGemini 3.1 Pro Preview
FrontierMath (Tiers 1-3)24.6%59.6%
FrontierMath Tier 40%26.8%
OTIS Mock AIME 2024-202584.7%95.6%
LMArena Math14501485
FrontierMath (Feb 2025 set)14.1%36.9%
FrontierMath Tier 4 (v1)4.2%16.7%
MathArena Final-Answer Competitions—86.5%
ProofBench—26%
Omni-MATH41.6%—
LiveBench Math90.2%—
MATH Level 595.9%—

Knowledge Gemini 3.1 Pro Preview leads

Gemini 2.5 Pro: 56.0 (#46), Gemini 3.1 Pro Preview: 71.8 (#3)

BenchmarkGemini 2.5 ProGemini 3.1 Pro Preview
GPQA Diamond85.3%94.4%
Humanity's Last Exam21.6%46.4%
Vectara Hallucination Rate7%10.4%
LMArena Expert14521485
SimpleQA Verified—73.5%
MMLU-Pro86.3%—
Confabulations10.6%—
GPQA (HELM)74.9%—

Multimodal Gemini 2.5 Pro leads

Gemini 2.5 Pro: 45.2 (#18), Gemini 3.1 Pro Preview: 37.9 (#69)

BenchmarkGemini 2.5 ProGemini 3.1 Pro Preview
LMArena Vision12631296
LMArena Document14211444
GeoBench86%—
VPCT48%—
Blueprint-Bench 2—26.5%
Furniture Assembly—26.7%
SpatialViz-Bench44.7%—

Multilingual Gemini 3.1 Pro Preview leads

Gemini 2.5 Pro: 55.3 (#31), Gemini 3.1 Pro Preview: 57.0 (#12)

BenchmarkGemini 2.5 ProGemini 3.1 Pro Preview
LMArena Non-English14511477
LMArena Chinese15071529
LMArena French14721487
LMArena German14871491
LMArena Japanese14611493
LMArena Korean14341455
LMArena Russian14611498
LMArena Spanish14731479

Instruction Following Gemini 3.1 Pro Preview leads

Gemini 2.5 Pro: 75.0 (#75), Gemini 3.1 Pro Preview: 77.0 (#32)

BenchmarkGemini 2.5 ProGemini 3.1 Pro Preview
LMArena Instruction Following14371466
LiveBench Instruction Following80.6%—
IFEval84%—

Long Context Gemini 2.5 Pro leads

Gemini 2.5 Pro: 59.8 (#5), Gemini 3.1 Pro Preview: 47.4 (#18)

BenchmarkGemini 2.5 ProGemini 3.1 Pro Preview
LMArena Longer Query14491483
Fiction.LiveBench91.7%—
CL-bench—20.8%
CL-bench Life—16.9%

Writing & Preference Gemini 3.1 Pro Preview leads

Gemini 2.5 Pro: 63.7 (#62), Gemini 3.1 Pro Preview: 66.1 (#37)

BenchmarkGemini 2.5 ProGemini 3.1 Pro Preview
LMArena Text14581481
LMArena Creative Writing14541482
EQ-Bench Creative Writing14211491
LMArena Multi-Turn14531488
Short-Story Creative Writing83.8%—
WildBench85.7%—
EQ-Bench 4—1142
LiveBench Language67.8%—

Frequently asked questions

Is Gemini 2.5 Pro better than Gemini 3.1 Pro Preview?

Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 45.0 on the Noometry Index.

Which is cheaper, Gemini 2.5 Pro or Gemini 3.1 Pro Preview?

Gemini 2.5 Pro is cheaper. It lists at $1.25 per million input tokens and $10 per million output tokens; Gemini 3.1 Pro Preview lists at $2 and $12.

Is Gemini 2.5 Pro or Gemini 3.1 Pro Preview better for coding?

They score almost the same on coding (42.4 vs 42.5); test both on your own repository before choosing.

Which has the bigger context window?

Both accept 1.05M tokens.

How many benchmarks do Gemini 2.5 Pro and Gemini 3.1 Pro Preview share?

52 benchmarks have published results for both models. Gemini 2.5 Pro has 78 scored results on Noometry and Gemini 3.1 Pro Preview has 71.

Related comparisons

Go deeper