Noometry

Model comparison

Gemma 3 12B vs Grok-3 mini

Grok-3 mini is the stronger model overall, scoring 41.0 to 31.8 on the Noometry Index.

Last verified . 15 shared benchmarks.

Gemma 3 12B Google

31.8

Rank #255 Confirmed

Grok-3 mini xAI

41.0

Rank #130 Confirmed

Summary

  • They share 15 benchmarks with published results for both. Gemma 3 12B scores higher in 1 category and Grok-3 mini in 7 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where Grok-3 mini leads 46.3 to 26.5.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 16.7% for Gemma 3 12B and 77.8% for Grok-3 mini.
  • Gemma 3 12B has downloadable open weights; the other is API-only.

Side by side

Gemma 3 12BGrok-3 mini
ProviderGooglexAI
Noometry Index31.841.0
Released2025-03-122025-04-09
WeightsOpenProprietary
Context window131K—
Max output8K—
Input $ / M tokens$0.05—
Output $ / M tokens$0.15—
Results tracked2435

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

Category by category

Coding Grok-3 mini leads

Gemma 3 12B: 31.2 (#276), Grok-3 mini: 40.4 (#126)

BenchmarkGemma 3 12BGrok-3 mini
LMArena Coding12811380
Aider Polyglot—49.3%
SciCode17.4%—
WeirdML—42.6%

Agentic & Tool Use Not comparable

Gemma 3 12B: 25.3 (#101), Grok-3 mini: —

BenchmarkGemma 3 12BGrok-3 mini
Berkeley Function Calling Leaderboard30.4%—

Reasoning Gemma 3 12B leads

Gemma 3 12B: 15.2 (#307), Grok-3 mini: 13.8 (#323)

BenchmarkGemma 3 12BGrok-3 mini
LMArena Hard Prompts13091374
Epoch Capabilities Index123.49140.35
ARC-AGI-2—0.4%
Kagi LLM Benchmark—61.3%
ARC-AGI-1—16.5%
CritPt0%—
Chess Puzzles0%—
DTBench48.8%—
LMCA4.5%—

Math Grok-3 mini leads

Gemma 3 12B: 21.0 (#273), Grok-3 mini: 40.3 (#78)

BenchmarkGemma 3 12BGrok-3 mini
OTIS Mock AIME 2024-202516.7%77.8%
LMArena Math13071386
Omni-MATH—31.8%
MATH Level 5—90.9%
FrontierMath (Feb 2025 set)—5.9%

Knowledge Grok-3 mini leads

Gemma 3 12B: 26.5 (#250), Grok-3 mini: 46.3 (#75)

BenchmarkGemma 3 12BGrok-3 mini
GPQA Diamond39.5%76.3%
LMArena Expert12481395
MMLU-Pro—79.9%
Confabulations—10.8%
Vectara Hallucination Rate4.4%—
GPQA (HELM)—67.5%

Multimodal Not comparable

Gemma 3 12B: —, Grok-3 mini: —

BenchmarkGemma 3 12BGrok-3 mini
MindCube46.7%—

Multilingual Grok-3 mini leads

Gemma 3 12B: 45.9 (#160), Grok-3 mini: 48.3 (#141)

BenchmarkGemma 3 12BGrok-3 mini
LMArena Non-English13181352
LMArena German13701350
LMArena Russian13351352
LMArena Chinese—1387
LMArena French—1357
LMArena Japanese—1343
LMArena Korean—1336
LMArena Spanish—1379

Instruction Following Grok-3 mini leads

Gemma 3 12B: 68.8 (#179), Grok-3 mini: 78.7 (#8)

BenchmarkGemma 3 12BGrok-3 mini
LMArena Instruction Following12991356
IFEval—95.1%

Long Context Too close to call

Gemma 3 12B: 40.0 (#157), Grok-3 mini: 41.0 (#142)

BenchmarkGemma 3 12BGrok-3 mini
LMArena Longer Query13171372
Fiction.LiveBench—66.7%

Writing & Preference Grok-3 mini leads

Gemma 3 12B: 47.7 (#205), Grok-3 mini: 52.7 (#163)

BenchmarkGemma 3 12BGrok-3 mini
LMArena Text13341370
LMArena Creative Writing13311342
LMArena Multi-Turn13341355
Short-Story Creative Writing—73.5%
EQ-Bench Creative Writing1126—
WildBench—65.1%

Frequently asked questions

Is Gemma 3 12B better than Grok-3 mini?

Grok-3 mini is the stronger model overall, scoring 41.0 to 31.8 on the Noometry Index.

Is Gemma 3 12B or Grok-3 mini better for coding?

Grok-3 mini scores higher on coding benchmarks: 40.4 versus 31.2 in the Noometry coding category.

How many benchmarks do Gemma 3 12B and Grok-3 mini share?

15 benchmarks have published results for both models. Gemma 3 12B has 24 scored results on Noometry and Grok-3 mini has 35.

Related comparisons

Go deeper