Noometry

Model comparison

Gemma 3 4B vs GPT-4.5

GPT-4.5 is the stronger model overall, scoring 37.2 to 28.1 on the Noometry Index.

Last verified . 16 shared benchmarks.

Gemma 3 4B Google

28.1

Rank #326 Confirmed

GPT-4.5 OpenAI

37.2

Rank #208 Confirmed

Summary

  • They share 16 benchmarks with published results for both. Gemma 3 4B scores higher in 0 categories and GPT-4.5 in 9 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where GPT-4.5 leads 32.5 to 11.8.
  • The biggest single-benchmark swing is GPQA Diamond: 23.2% for Gemma 3 4B and 68.7% for GPT-4.5.
  • Gemma 3 4B has downloadable open weights; the other is API-only.

Side by side

Gemma 3 4BGPT-4.5
ProviderGoogleOpenAI
Noometry Index28.137.2
Released2025-03-122025-02-27
WeightsOpenProprietary
Context window131K—
Max output4K—
Input $ / M tokens$0.04—
Output $ / M tokens$0.08—
Results tracked2242

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

Category by category

Coding GPT-4.5 leads

Gemma 3 4B: 35.9 (#215), GPT-4.5: 42.2 (#109)

BenchmarkGemma 3 4BGPT-4.5
LMArena Coding12301396
Aider Polyglot—44.9%
WeirdML—39.4%
LiveBench Coding—75.2%

Agentic & Tool Use GPT-4.5 leads

Gemma 3 4B: 20.9 (#142), GPT-4.5: 27.9 (#97)

BenchmarkGemma 3 4BGPT-4.5
Berkeley Function Calling Leaderboard19.6%—
Cybench—17.5%

Reasoning Too close to call

Gemma 3 4B: 13.2 (#335), GPT-4.5: 13.9 (#330)

BenchmarkGemma 3 4BGPT-4.5
LMArena Hard Prompts12531403
Epoch Capabilities Index116.02136.74
ARC-AGI-2—0.8%
SimpleBench—34.5%
Kagi LLM Benchmark25.2%—
ARC-AGI-1—10.3%
Chess Puzzles0%—
EnigmaEval—3.2%
LiveBench Reasoning—71.1%
DTBench50.9%—
LiveBench Data Analysis—64.3%
LMCA2.8%—
ForecastBench—61.7
LiveBench—69%

Math GPT-4.5 leads

Gemma 3 4B: 16.8 (#292), GPT-4.5: 32.6 (#211)

BenchmarkGemma 3 4BGPT-4.5
OTIS Mock AIME 2024-20257.5%37.8%
LMArena Math12391412
LiveBench Math—69.3%
MATH Level 5—78.6%

Knowledge GPT-4.5 leads

Gemma 3 4B: 11.8 (#299), GPT-4.5: 32.5 (#211)

BenchmarkGemma 3 4BGPT-4.5
GPQA Diamond23.2%68.7%
LMArena Expert12231394
Humanity's Last Exam—5.4%
Confabulations—13.6%
Vectara Hallucination Rate6.4%—

Multimodal Not comparable

Gemma 3 4B: —, GPT-4.5: 37.6 (#71)

BenchmarkGemma 3 4BGPT-4.5
LMArena Vision—1195
VPCT—45%

Multilingual GPT-4.5 leads

Gemma 3 4B: 42.5 (#194), GPT-4.5: 52.5 (#83)

BenchmarkGemma 3 4BGPT-4.5
LMArena Non-English12731413
LMArena German12811457
LMArena Russian12941419
LMArena Chinese—1421
LMArena French—1418
LMArena Japanese—1416
LMArena Korean—1392

Instruction Following GPT-4.5 leads

Gemma 3 4B: 65.2 (#225), GPT-4.5: 72.6 (#134)

BenchmarkGemma 3 4BGPT-4.5
LMArena Instruction Following12391404
LiveBench Instruction Following—72.3%

Long Context GPT-4.5 leads

Gemma 3 4B: 38.7 (#194), GPT-4.5: 40.4 (#155)

BenchmarkGemma 3 4BGPT-4.5
LMArena Longer Query12731406
Fiction.LiveBench—63.9%

Writing & Preference GPT-4.5 leads

Gemma 3 4B: 42.0 (#239), GPT-4.5: 56.9 (#134)

BenchmarkGemma 3 4BGPT-4.5
LMArena Text12911417
LMArena Creative Writing12711394
EQ-Bench Creative Writing10681258
LMArena Multi-Turn12551444
Short-Story Creative Writing—75.6%
LiveBench Language—61.5%

Frequently asked questions

Is Gemma 3 4B better than GPT-4.5?

GPT-4.5 is the stronger model overall, scoring 37.2 to 28.1 on the Noometry Index.

Is Gemma 3 4B or GPT-4.5 better for coding?

GPT-4.5 scores higher on coding benchmarks: 42.2 versus 35.9 in the Noometry coding category.

How many benchmarks do Gemma 3 4B and GPT-4.5 share?

16 benchmarks have published results for both models. Gemma 3 4B has 22 scored results on Noometry and GPT-4.5 has 42.

Related comparisons

Go deeper