Noometry

Model comparison

Claude 3 Haiku vs Gemma 3 4B

Gemma 3 4B is the stronger model overall, scoring 28.1 to 25.9 on the Noometry Index.

Last verified . 19 shared benchmarks.

Claude 3 Haiku Anthropic

25.9

Rank #340 Confirmed

Gemma 3 4B Google

28.1

Rank #326 Confirmed

Summary

  • They share 19 benchmarks with published results for both. Claude 3 Haiku scores higher in 2 categories and Gemma 3 4B in 6 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where Gemma 3 4B leads 42.0 to 29.7.
  • The biggest single-benchmark swing is GPQA Diamond: 36.3% for Claude 3 Haiku and 23.2% for Gemma 3 4B.
  • Gemma 3 4B has downloadable open weights; the other is API-only.

Side by side

Claude 3 HaikuGemma 3 4B
ProviderAnthropicGoogle
Noometry Index25.928.1
Released2024-03-072025-03-12
WeightsProprietaryOpen
Context window—131K
Max output—4K
Input $ / M tokens—$0.04
Output $ / M tokens—$0.08
Results tracked3722

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

Category by category

Coding Gemma 3 4B leads

Claude 3 Haiku: 26.4 (#325), Gemma 3 4B: 35.9 (#215)

BenchmarkClaude 3 HaikuGemma 3 4B
LMArena Coding11991230
WeirdML9.8%—
BigCodeBench Instruct39.4%—
BigCodeBench Complete50.1%—
CadEval12%—
HumanEval+68.9%—
MBPP+68.8%—

Agentic & Tool Use Not comparable

Claude 3 Haiku: —, Gemma 3 4B: 20.9 (#142)

BenchmarkClaude 3 HaikuGemma 3 4B
Berkeley Function Calling Leaderboard—19.6%

Reasoning Claude 3 Haiku leads

Claude 3 Haiku: 16.3 (#307), Gemma 3 4B: 13.2 (#335)

BenchmarkClaude 3 HaikuGemma 3 4B
Kagi LLM Benchmark34.2%25.2%
LMArena Hard Prompts11741253
DTBench50.1%50.9%
LMCA8.8%2.8%
Epoch Capabilities Index118.35116.02
Chess Puzzles—0%
ForecastBench53.2—
WinoGrande74.2%—

Math Gemma 3 4B leads

Claude 3 Haiku: 9.8 (#319), Gemma 3 4B: 16.8 (#292)

BenchmarkClaude 3 HaikuGemma 3 4B
OTIS Mock AIME 2024-20251.8%7.5%
LMArena Math11881239
MATH Level 514.9%—

Knowledge Claude 3 Haiku leads

Claude 3 Haiku: 17.3 (#285), Gemma 3 4B: 11.8 (#299)

BenchmarkClaude 3 HaikuGemma 3 4B
GPQA Diamond36.3%23.2%
LMArena Expert11481223
Confabulations34.2%—
Vectara Hallucination Rate—6.4%
MMLU73.8%—

Multimodal Not comparable

Claude 3 Haiku: 23.6 (#128), Gemma 3 4B: —

BenchmarkClaude 3 HaikuGemma 3 4B
LMArena Vision950—
ScienceQA72%—

Multilingual Gemma 3 4B leads

Claude 3 Haiku: 36.0 (#243), Gemma 3 4B: 42.5 (#194)

BenchmarkClaude 3 HaikuGemma 3 4B
LMArena Non-English11781273
LMArena German11741281
LMArena Russian12041294
LMArena Chinese1155—
LMArena French1195—
LMArena Japanese1102—
LMArena Korean1109—
LMArena Spanish1166—

Instruction Following Gemma 3 4B leads

Claude 3 Haiku: 61.3 (#247), Gemma 3 4B: 65.2 (#225)

BenchmarkClaude 3 HaikuGemma 3 4B
LMArena Instruction Following11731239

Long Context Gemma 3 4B leads

Claude 3 Haiku: 36.1 (#237), Gemma 3 4B: 38.7 (#194)

BenchmarkClaude 3 HaikuGemma 3 4B
LMArena Longer Query11901273

Writing & Preference Gemma 3 4B leads

Claude 3 Haiku: 29.7 (#291), Gemma 3 4B: 42.0 (#239)

BenchmarkClaude 3 HaikuGemma 3 4B
LMArena Text11951291
LMArena Creative Writing11571271
EQ-Bench Creative Writing7171068
LMArena Multi-Turn11901255

Frequently asked questions

Is Claude 3 Haiku better than Gemma 3 4B?

Gemma 3 4B is the stronger model overall, scoring 28.1 to 25.9 on the Noometry Index.

Is Claude 3 Haiku or Gemma 3 4B better for coding?

Gemma 3 4B scores higher on coding benchmarks: 35.9 versus 26.4 in the Noometry coding category.

How many benchmarks do Claude 3 Haiku and Gemma 3 4B share?

19 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Gemma 3 4B has 22.

Related comparisons

Go deeper