Noometry

Model comparison

Gemini 3.1 Pro Preview vs GPT-5.5

GPT-5.5 is the stronger model overall, scoring 63.4 to 56.7 on the Noometry Index. Gemini 3.1 Pro Preview costs 2.5× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.

Last verified . 64 shared benchmarks.

Gemini 3.1 Pro Preview Google

56.7

Rank #23 Confirmed

GPT-5.5 OpenAI

63.4

Rank #9 Confirmed

Summary

  • They share 64 benchmarks with published results for both. Gemini 3.1 Pro Preview scores higher in 2 categories and GPT-5.5 in 8 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in math, where GPT-5.5 leads 81.7 to 62.1.
  • The biggest single-benchmark swing is DeepSWE: 11.7% for Gemini 3.1 Pro Preview and 67% for GPT-5.5.
  • Gemini 3.1 Pro Preview is cheaper at $2 / $12 per million input/output tokens, against $5 / $30 for GPT-5.5.
  • GPT-5.5 accepts more context: 1.05M tokens versus 1.05M.

Side by side

Gemini 3.1 Pro PreviewGPT-5.5
ProviderGoogleOpenAI
Noometry Index56.763.4
Released2026-02-192026-04-23
WeightsProprietaryProprietary
Context window1.05M1.05M
Max output66K128K
Input $ / M tokens$2$5
Output $ / M tokens$12$30
Results tracked7171

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

Category by category

Coding GPT-5.5 leads

Gemini 3.1 Pro Preview: 42.5 (#99), GPT-5.5: 58.2 (#17)

BenchmarkGemini 3.1 Pro PreviewGPT-5.5
SWE-bench Verified75.6%80.6%
DeepSWE11.7%67%
LMArena WebDev14471513
SciCode58.9%56.1%
GSO22.6%40.2%
WeirdML72.1%84.9%
LMArena Coding14841494
MirrorCode8.9%10%
ALE-Bench1,1611,943
FrontierCode—43%
AlgoTune2.02—

Agentic & Tool Use GPT-5.5 leads

Gemini 3.1 Pro Preview: 37.7 (#34), GPT-5.5: 50.7 (#6)

BenchmarkGemini 3.1 Pro PreviewGPT-5.5
Terminal-Bench80.2%84.7%
APEX-Agents35.3%55.1%
τ²-bench Banking26%44.6%
DeepResearch Bench47.8%54%
PostTrainBench22%27.2%
ExploitBench26.1%47.4%
GBAEval0.8%53.2%
GDP.pdf17%26%
LMArena Search12111242
Vending-Bench 23,7747,524
OSWorld 2.0—13%
Remote Labor Index—6.3%
BALROG57%—
METR Time Horizons77%—

Reasoning GPT-5.5 leads

Gemini 3.1 Pro Preview: 71.7 (#12), GPT-5.5: 72.8 (#11)

BenchmarkGemini 3.1 Pro PreviewGPT-5.5
ARC-AGI-277.1%85%
SimpleBench79.6%69%
NYT Connections (extended)97.4%96.2%
ARC-AGI-198%95%
CritPt17.7%27.1%
Chess Puzzles55%54%
EBR-Bench14.3%34.3%
LMArena Hard Prompts14851489
Mystery Game Puzzles34%56%
DTBench97.1%96%
LMCA53.8%54.3%
Epoch Capabilities Index154.77159.1
ForecastBench5960.6
Kagi LLM Benchmark—88.8%
EnigmaEval36.8%—
Thematic Generalization79.4%—
Surface Evolver Bench—88.1%
Bench to the Future 3—0.14

Math GPT-5.5 leads

Gemini 3.1 Pro Preview: 62.1 (#34), GPT-5.5: 81.7 (#11)

BenchmarkGemini 3.1 Pro PreviewGPT-5.5
FrontierMath (Tiers 1-3)59.6%85.3%
FrontierMath Tier 426.8%72.5%
MathArena Final-Answer Competitions86.5%94.3%
OTIS Mock AIME 2024-202595.6%100%
ProofBench26%50%
LMArena Math14851486
FrontierMath (Feb 2025 set)36.9%51.7%
FrontierMath Tier 4 (v1)16.7%35.4%
FrontierMath Erdős—0%

Knowledge Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 71.8 (#3), GPT-5.5: 64.4 (#17)

BenchmarkGemini 3.1 Pro PreviewGPT-5.5
GPQA Diamond94.4%94%
SimpleQA Verified73.5%63%
Vectara Hallucination Rate10.4%9.3%
LMArena Expert14851508
Humanity's Last Exam46.4%—

Multimodal GPT-5.5 leads

Gemini 3.1 Pro Preview: 37.9 (#69), GPT-5.5: 46.9 (#12)

BenchmarkGemini 3.1 Pro PreviewGPT-5.5
LMArena Vision12961297
Blueprint-Bench 226.5%36.2%
Furniture Assembly26.7%44.2%
LMArena Document14441486

Multilingual Too close to call

Gemini 3.1 Pro Preview: 57.0 (#12), GPT-5.5: 56.4 (#20)

BenchmarkGemini 3.1 Pro PreviewGPT-5.5
LMArena Non-English14771467
LMArena Chinese15291533
LMArena French14871486
LMArena German14911480
LMArena Japanese14931498
LMArena Korean14551460
LMArena Russian14981473
LMArena Spanish14791468

Instruction Following Too close to call

Gemini 3.1 Pro Preview: 77.0 (#32), GPT-5.5: 77.5 (#18)

BenchmarkGemini 3.1 Pro PreviewGPT-5.5
LMArena Instruction Following14661479

Long Context Too close to call

Gemini 3.1 Pro Preview: 47.4 (#18), GPT-5.5: 48.3 (#12)

BenchmarkGemini 3.1 Pro PreviewGPT-5.5
CL-bench Life16.9%22.2%
LMArena Longer Query14831484
CL-bench20.8%—

Writing & Preference GPT-5.5 leads

Gemini 3.1 Pro Preview: 66.1 (#37), GPT-5.5: 72.7 (#13)

BenchmarkGemini 3.1 Pro PreviewGPT-5.5
LMArena Text14811472
LMArena Creative Writing14821455
EQ-Bench Creative Writing14911844
EQ-Bench 411421315
LMArena Multi-Turn14881476

Frequently asked questions

Is Gemini 3.1 Pro Preview better than GPT-5.5?

GPT-5.5 is the stronger model overall, scoring 63.4 to 56.7 on the Noometry Index. Gemini 3.1 Pro Preview costs 2.5× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.

Which is cheaper, Gemini 3.1 Pro Preview or GPT-5.5?

Gemini 3.1 Pro Preview is cheaper. It lists at $2 per million input tokens and $12 per million output tokens; GPT-5.5 lists at $5 and $30.

Is Gemini 3.1 Pro Preview or GPT-5.5 better for coding?

GPT-5.5 scores higher on coding benchmarks: 58.2 versus 42.5 in the Noometry coding category.

Which has the bigger context window?

GPT-5.5 does, with 1.05M tokens against 1.05M.

How many benchmarks do Gemini 3.1 Pro Preview and GPT-5.5 share?

64 benchmarks have published results for both models. Gemini 3.1 Pro Preview has 71 scored results on Noometry and GPT-5.5 has 71.

Related comparisons

Go deeper