Noometry

Model comparison

Gemini 3.1 Pro Preview vs GPT-5

Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 50.9 on the Noometry Index.

Last verified . 53 shared benchmarks.

Gemini 3.1 Pro Preview Google

56.7

Rank #23 Confirmed

GPT-5 OpenAI

50.9

Rank #45 Confirmed

Summary

  • They share 53 benchmarks with published results for both. Gemini 3.1 Pro Preview scores higher in 7 categories and GPT-5 in 3 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Gemini 3.1 Pro Preview leads 71.7 to 38.3.
  • The biggest single-benchmark swing is ARC-AGI-2: 77.1% for Gemini 3.1 Pro Preview and 9.9% for GPT-5.
  • GPT-5 is cheaper at $1.25 / $10 per million input/output tokens, against $2 / $12 for Gemini 3.1 Pro Preview.
  • Gemini 3.1 Pro Preview accepts more context: 1.05M tokens versus 400K.

Side by side

Gemini 3.1 Pro PreviewGPT-5
ProviderGoogleOpenAI
Noometry Index56.750.9
Released2026-02-192025-08-07
WeightsProprietaryProprietary
Context window1.05M400K
Max output66K128K
Input $ / M tokens$2$1.25
Output $ / M tokens$12$10
Results tracked7169

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

Category by category

Coding GPT-5 leads

Gemini 3.1 Pro Preview: 42.5 (#99), GPT-5: 50.3 (#47)

BenchmarkGemini 3.1 Pro PreviewGPT-5
SWE-bench Verified75.6%73.6%
LMArena WebDev14471418
SciCode58.9%42.9%
GSO22.6%6.9%
WeirdML72.1%60.7%
LMArena Coding14841436
ALE-Bench1,1611,162
AlgoTune2.021.67
DeepSWE11.7%—
SWE-bench Verified (bash only)—65%
Aider Polyglot—88%
MirrorCode8.9%—

Agentic & Tool Use Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 37.7 (#34), GPT-5: 33.1 (#56)

BenchmarkGemini 3.1 Pro PreviewGPT-5
Terminal-Bench80.2%49.6%
DeepResearch Bench47.8%49.6%
BALROG57%32.8%
LMArena Search12111133
METR Time Horizons77%69.6%
APEX-Agents35.3%—
GDPval—34.8%
Remote Labor Index—1.7%
τ²-bench Banking26%—
PostTrainBench22%—
ExploitBench26.1%—
GBAEval0.8%—
GDP.pdf17%—
Vending-Bench 23,774—

Reasoning Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 71.7 (#12), GPT-5: 38.3 (#64)

BenchmarkGemini 3.1 Pro PreviewGPT-5
ARC-AGI-277.1%9.9%
SimpleBench79.6%56.7%
ARC-AGI-198%65.7%
CritPt17.7%12.6%
Chess Puzzles55%37%
EnigmaEval36.8%10.5%
EBR-Bench14.3%12.7%
LMArena Hard Prompts14851416
Mystery Game Puzzles34%23%
DTBench97.1%90.7%
LMCA53.8%40%
Epoch Capabilities Index154.77150
ForecastBench5961.4
Kagi LLM Benchmark—72.7%
NYT Connections (extended)97.4%—
Thematic Generalization79.4%—

Math Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 62.1 (#34), GPT-5: 55.0 (#44)

BenchmarkGemini 3.1 Pro PreviewGPT-5
FrontierMath (Tiers 1-3)59.6%55.4%
FrontierMath Tier 426.8%22%
OTIS Mock AIME 2024-202595.6%91.4%
ProofBench26%18%
LMArena Math14851407
FrontierMath (Feb 2025 set)36.9%32.4%
FrontierMath Tier 4 (v1)16.7%12.5%
MathArena Final-Answer Competitions86.5%—
Omni-MATH—64.7%
MATH Level 5—98.1%

Knowledge Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 71.8 (#3), GPT-5: 56.6 (#43)

BenchmarkGemini 3.1 Pro PreviewGPT-5
GPQA Diamond94.4%86.2%
Humanity's Last Exam46.4%25.3%
SimpleQA Verified73.5%50.1%
Vectara Hallucination Rate10.4%14.7%
LMArena Expert14851419
MMLU-Pro—86.3%
Confabulations—10.3%
GPQA (HELM)—79.2%

Multimodal GPT-5 leads

Gemini 3.1 Pro Preview: 37.9 (#69), GPT-5: 46.8 (#13)

BenchmarkGemini 3.1 Pro PreviewGPT-5
LMArena Vision12961232
GeoBench—81%
VPCT—66%
Blueprint-Bench 226.5%—
Furniture Assembly26.7%—
LMArena Document1444—

Multilingual Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 57.0 (#12), GPT-5: 51.4 (#110)

BenchmarkGemini 3.1 Pro PreviewGPT-5
LMArena Non-English14771397
LMArena Chinese15291422
LMArena French14871410
LMArena German14911416
LMArena Japanese14931409
LMArena Korean14551360
LMArena Russian14981406
LMArena Spanish14791399

Instruction Following Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 77.0 (#32), GPT-5: 73.8 (#113)

BenchmarkGemini 3.1 Pro PreviewGPT-5
LMArena Instruction Following14661388
IFEval—87.5%

Long Context GPT-5 leads

Gemini 3.1 Pro Preview: 47.4 (#18), GPT-5: 69.5 (#2)

BenchmarkGemini 3.1 Pro PreviewGPT-5
LMArena Longer Query14831399
Fiction.LiveBench—97.2%
CL-bench20.8%—
CL-bench Life16.9%—

Writing & Preference Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 66.1 (#37), GPT-5: 63.4 (#65)

BenchmarkGemini 3.1 Pro PreviewGPT-5
LMArena Text14811406
LMArena Creative Writing14821365
EQ-Bench Creative Writing14911627
LMArena Multi-Turn14881426
Short-Story Creative Writing—86%
WildBench—85.7%
EQ-Bench 41142—

Frequently asked questions

Is Gemini 3.1 Pro Preview better than GPT-5?

Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 50.9 on the Noometry Index.

Which is cheaper, Gemini 3.1 Pro Preview or GPT-5?

GPT-5 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 3.1 Pro Preview or GPT-5 better for coding?

GPT-5 scores higher on coding benchmarks: 50.3 versus 42.5 in the Noometry coding category.

Which has the bigger context window?

Gemini 3.1 Pro Preview does, with 1.05M tokens against 400K.

How many benchmarks do Gemini 3.1 Pro Preview and GPT-5 share?

53 benchmarks have published results for both models. Gemini 3.1 Pro Preview has 71 scored results on Noometry and GPT-5 has 69.

Related comparisons

Go deeper