Noometry

Model comparison

Gemini 3.1 Pro Preview vs GPT-5.2

Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 54.1 on the Noometry Index.

Last verified . 57 shared benchmarks.

Gemini 3.1 Pro Preview Google

56.7

Rank #20 Confirmed

GPT-5.2 OpenAI

54.1

Rank #31 Confirmed

Summary

  • They share 57 benchmarks with published results for both. Gemini 3.1 Pro Preview scores higher in 6 categories and GPT-5.2 in 4 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Gemini 3.1 Pro Preview leads 71.7 to 50.1.
  • The biggest single-benchmark swing is SimpleQA Verified: 73.5% for Gemini 3.1 Pro Preview and 37.1% for GPT-5.2.
  • Gemini 3.1 Pro Preview is cheaper at $2 / $12 per million input/output tokens, against $1.75 / $14 for GPT-5.2.
  • Gemini 3.1 Pro Preview accepts more context: 1.05M tokens versus 400K.

Side by side

Gemini 3.1 Pro PreviewGPT-5.2
ProviderGoogleOpenAI
Noometry Index56.754.1
Released2026-02-192025-12-11
WeightsProprietaryProprietary
Context window1.05M400K
Max output66K128K
Input $ / M tokens$2$1.75
Output $ / M tokens$12$14
Results tracked7167

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

Category by category

Coding GPT-5.2 leads

Gemini 3.1 Pro Preview: 42.7 (#88), GPT-5.2: 52.2 (#32)

BenchmarkGemini 3.1 Pro PreviewGPT-5.2
SWE-bench Verified75.6%73.8%
LMArena WebDev14461415
GSO22.6%27.4%
WeirdML72.1%72.2%
LMArena Coding14831447
ALE-Bench1,1611,294
AlgoTune2.022.05
DeepSWE11.7%—
SWE-bench Verified (bash only)—72.8%
SWE-bench Multilingual—66.7%
SciCode58.9%—
MirrorCode8.9%—

Agentic & Tool Use GPT-5.2 leads

Gemini 3.1 Pro Preview: 37.7 (#33), GPT-5.2: 40.0 (#24)

BenchmarkGemini 3.1 Pro PreviewGPT-5.2
Terminal-Bench80.2%64.9%
τ²-bench Banking26%32.2%
DeepResearch Bench47.8%41.1%
LMArena Search12111207
METR Time Horizons77%75.3%
Vending-Bench 23,7743,591
APEX-Agents35.3%—
Berkeley Function Calling Leaderboard—55.9%
GDPval—49.7%
Remote Labor Index—2.5%
τ²-bench Airline—83%
τ²-bench Retail—81.6%
τ²-bench Telecom—89.7%
PostTrainBench22%—
BALROG57%—
ExploitBench26.1%—
GBAEval0.8%—
GDP.pdf17%—

Reasoning Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 71.7 (#9), GPT-5.2: 50.1 (#33)

BenchmarkGemini 3.1 Pro PreviewGPT-5.2
ARC-AGI-277.1%52.9%
SimpleBench79.6%45.8%
NYT Connections (extended)97.4%83.6%
ARC-AGI-198%86.2%
Chess Puzzles55%49%
EnigmaEval36.8%10.4%
EBR-Bench14.3%23%
LMArena Hard Prompts14841445
Mystery Game Puzzles34%23%
DTBench97.1%90.9%
LMCA53.8%43.9%
Epoch Capabilities Index154.92153.45
ForecastBench5960.1
Kagi LLM Benchmark—73.3%
CritPt17.7%—
Thematic Generalization79.4%—

Math Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 61.8 (#28), GPT-5.2: 59.8 (#32)

BenchmarkGemini 3.1 Pro PreviewGPT-5.2
FrontierMath (Tiers 1-3)59.6%67.4%
FrontierMath Tier 426.8%31.7%
MathArena Final-Answer Competitions86.5%72%
OTIS Mock AIME 2024-202595.6%96.1%
ProofBench26%15%
LMArena Math14851439
FrontierMath (Feb 2025 set)36.9%40.7%
FrontierMath Tier 4 (v1)16.7%18.8%

Knowledge Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 71.8 (#3), GPT-5.2: 59.2 (#29)

BenchmarkGemini 3.1 Pro PreviewGPT-5.2
GPQA Diamond94.4%91.4%
Humanity's Last Exam46.4%27.8%
SimpleQA Verified73.5%37.1%
Vectara Hallucination Rate10.4%8.4%
LMArena Expert14841445

Multimodal GPT-5.2 leads

Gemini 3.1 Pro Preview: 38.1 (#60), GPT-5.2: 50.9 (#7)

BenchmarkGemini 3.1 Pro PreviewGPT-5.2
LMArena Vision12961268
Furniture Assembly26.7%38.3%
LMArena Document14441405
VPCT—84%
Blueprint-Bench 226.5%—

Multilingual Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 57.1 (#11), GPT-5.2: 53.6 (#63)

BenchmarkGemini 3.1 Pro PreviewGPT-5.2
LMArena Non-English14751425
LMArena Chinese15281460
LMArena French14881456
LMArena German14911448
LMArena Japanese14941421
LMArena Korean14581390
LMArena Russian14961440
LMArena Spanish14801433

Instruction Following Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 77.1 (#31), GPT-5.2: 74.9 (#83)

BenchmarkGemini 3.1 Pro PreviewGPT-5.2
LMArena Instruction Following14651417

Long Context Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 47.3 (#17), GPT-5.2: 44.0 (#74)

BenchmarkGemini 3.1 Pro PreviewGPT-5.2
CL-bench20.8%18.2%
LMArena Longer Query14821428
CL-bench Life16.9%—

Writing & Preference Too close to call

Gemini 3.1 Pro Preview: 66.3 (#36), GPT-5.2: 67.0 (#30)

BenchmarkGemini 3.1 Pro PreviewGPT-5.2
LMArena Text14801439
LMArena Creative Writing14821402
EQ-Bench Creative Writing14911703
LMArena Multi-Turn14861458
EQ-Bench 41142—

Frequently asked questions

Is Gemini 3.1 Pro Preview better than GPT-5.2?

Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 54.1 on the Noometry Index.

Which is cheaper, Gemini 3.1 Pro Preview or GPT-5.2?

Gemini 3.1 Pro Preview is cheaper. It lists at $2 per million input tokens and $12 per million output tokens; GPT-5.2 lists at $1.75 and $14.

Is Gemini 3.1 Pro Preview or GPT-5.2 better for coding?

GPT-5.2 scores higher on coding benchmarks: 52.2 versus 42.7 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-5.2 share?

57 benchmarks have published results for both models. Gemini 3.1 Pro Preview has 71 scored results on Noometry and GPT-5.2 has 67.

Related comparisons

Go deeper