Skip to content
VECTOR WIREAI INTELLIGENCE
UTC

Gemini 3 Pro vs Seed2.1

16 SHARED BENCHMARKS

Across 16 shared benchmarks, Gemini 3 Pro scores higher on 2 and Seed2.1 on 14. The widest gap is GDPVal, where Seed2.1 scores 82.7 against 34.2.

GOOGLEVSX16 SHARED2–14 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 16 shared benchmarks2 wins14 wins
Scores tracked independently verified194 41 ◆50 0 ◆

Gemini 3 Pro's release date per Artificial Analysis. Prices: Artificial Analysis for Gemini 3 Pro. ◆ = independently verified score.

Where each leadsby capability area · benchmark wins

AreaGemini 3 ProWINSSeed2.1
Reasoning01Seed2.1 leads 1 of 1 · widest: ARC-AGI-2 31.1 vs 61.3
Coding01Seed2.1 leads 1 of 1
Agentic03Seed2.1 leads 3 of 3 · widest: GDPVal 34.2 vs 82.7
Multimodal22Even, 2–2 of 4

Biggest gaps

Gemini 3 Pro pulls furthest ahead on

No ratified-area lead of 3 points or more.

Seed2.1 pulls furthest ahead on

  1. GDPVal82.7 vs 34.2
  2. ARC-AGI-261.3 vs 31.1
  3. AA ApexAgents29.2 vs 18.4

Every shared benchmark16 · grouped by area

Reasoning 1

BenchmarkGemini 3 ProMARGINSeed2.1
ARC-AGI-231.161.3

Coding 1

BenchmarkGemini 3 ProMARGINSeed2.1
SciCode56.157.8

Agentic 3

BenchmarkGemini 3 ProMARGINSeed2.1
GDPVal34.282.7
MCP Atlas54.180.3

Multimodal 4

BenchmarkGemini 3 ProMARGINSeed2.1
MathVista89.890.5
MMMU-Pro80.280.1
OCRBenchv263.4 ◆62.8
Video-MME88.489

Other shared benchmarks 7

BenchmarkGemini 3 ProMARGINSeed2.1
LVBench73.576.8
MotionBench70.374.8
SimpleVQA69.771.1
Toolathlon36.449.1
WorldVQA47.448.6

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 Pro withALL PAIRINGS →

Compare Seed2.1 withALL PAIRINGS →