Skip to content
VECTOR WIREAI INTELLIGENCE
UTC

Gemini 3.1 Pro vs Seed2.1

42 SHARED BENCHMARKS

Across 42 shared benchmarks, Gemini 3.1 Pro scores higher on 12 and Seed2.1 on 29, with 1 level. The widest gap is GDPVal, where Seed2.1 scores 82.7 against 13.8.

GOOGLEVSX42 SHARED12–29 HEAD-TO-HEADNEWEST SCORE ADDED

At a glance

MakerGooglex
Released—
Price per 1M tokens input / output$2.00 / $12.00—
Cost of 1M in + 1M out$14.00—
Head-to-head of 42 shared benchmarks12 wins29 wins
Scores tracked independently verified215 30 ◆50 0 ◆

Gemini 3.1 Pro's release date per Artificial Analysis. Prices: Artificial Analysis for Gemini 3.1 Pro. ◆ = independently verified score.

Where each leadsby capability area · benchmark wins

AreaGemini 3.1 ProWINSSeed2.1
Reasoning10Gemini 3.1 Pro leads 1 of 1 · widest: ARC-AGI-2 77.1 vs 61.3
Coding20Gemini 3.1 Pro leads 2 of 2 · widest: Terminal-Bench 2.1 73.8 vs 67.6
Agentic12Seed2.1 leads 2 of 3 · widest: GDPVal 13.8 vs 82.7
Multimodal13Seed2.1 leads 3 of 5

Biggest gaps

Gemini 3.1 Pro pulls furthest ahead on

  1. ARC-AGI-277.1 vs 61.3
  2. Terminal-Bench 2.173.8 vs 67.6

Seed2.1 pulls furthest ahead on

  1. GDPVal82.7 vs 13.8
  2. MCP Atlas80.3 vs 69.2

Every shared benchmark42 · grouped by area

Reasoning 1

BenchmarkGemini 3.1 ProMARGINSeed2.1
ARC-AGI-277.161.3

Coding 2

BenchmarkGemini 3.1 ProMARGINSeed2.1
SciCode58.757.8

Agentic 3

BenchmarkGemini 3.1 ProMARGINSeed2.1
GDPVal13.882.7
MCP Atlas69.280.3

Multimodal 5

BenchmarkGemini 3.1 ProMARGINSeed2.1
BLINK79.179.4
MathVista90.290.5
MMMU-Pro82.480.1
OCRBenchv262.862.8
Video-MME86.789

Other shared benchmarks 31

BenchmarkGemini 3.1 ProMARGINSeed2.1
BabyVision54.462.9
ChartQAPro70.270.9
CyberGym38.867
DUDE82.183.1
DynaMath72.168.1
ERQA70.871.3
Finance Agent v1.159.756
LVBench66.276.8
MathVerse (Vision-Only)87.789.2
Minerva63.565.9
MMSIBench (circular)27.431.4
MotionBench69.974.8
Office QA Pro [Multimodal]72.571.1
OfficeQA Pro72.562.8
OVBench58.869.7
RealWorldQA85.486.3
SimpleVQA69.971.1
SWE-Pro Bench54.257
TOMATO60.456.8
Toolathlon48.849.1
TVBench7177.2
VideoHolmes65.967.6
VisuLogic44.852.9
WorldVQA44.348.6
ZeroBench (main)1211
ZeroBench (sub)41.949.1

Each score is the one the model's own page shows — the most authoritative tracked result for that benchmark, on the benchmark's standard methodology (source-attributed). ↓ marks lower-is-better metrics; ◆ an independently verified score. Open either model for its full surface, provenance and pricing.

Compare Gemini 3.1 Pro withALL PAIRINGS →

Compare Seed2.1 withALL PAIRINGS →