Noometry

Model comparison

Claude 3 Haiku vs Gemma 4 31B IT

Gemma 4 31B IT is the stronger model overall, scoring 43.2 to 25.7 on the Noometry Index.

Last verified . 23 shared benchmarks.

Claude 3 Haiku Anthropic

25.7

Rank #333 Confirmed

Gemma 4 31B IT Google

43.2

Rank #85 Confirmed

Summary

  • They share 23 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and Gemma 4 31B IT in 9 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in math, where Gemma 4 31B IT leads 41.3 to 9.2.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.8% for Claude 3 Haiku and 73.3% for Gemma 4 31B IT.
  • Gemma 4 31B IT has downloadable open weights; the other is API-only.

Side by side

Claude 3 HaikuGemma 4 31B IT
ProviderAnthropicGoogle
Noometry Index25.743.2
Released2024-03-072026-04-02
WeightsProprietaryOpen
Context window—262K
Max output—33K
Input $ / M tokens—$0.09
Output $ / M tokens—$0.34
Results tracked3735

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

Category by category

Coding Gemma 4 31B IT leads

Claude 3 Haiku: 26.1 (#320), Gemma 4 31B IT: 41.9 (#98)

BenchmarkClaude 3 HaikuGemma 4 31B IT
WeirdML9.8%52.3%
LMArena Coding11991459
LMArena WebDev—1364
SciCode—43.4%
BigCodeBench Instruct39.4%—
BigCodeBench Complete50.1%—
CadEval12%—
ALE-Bench—925.5
HumanEval+68.9%—
MBPP+68.8%—

Reasoning Gemma 4 31B IT leads

Claude 3 Haiku: 15.7 (#303), Gemma 4 31B IT: 26.4 (#119)

BenchmarkClaude 3 HaikuGemma 4 31B IT
Kagi LLM Benchmark34.2%63.5%
LMArena Hard Prompts11741448
DTBench50.1%82.7%
LMCA8.8%39.3%
Epoch Capabilities Index118.3142.74
NYT Connections (extended)—70.6%
CritPt—1.4%
Chess Puzzles—5%
Thematic Generalization—53%
Surface Evolver Bench—30.6%
ForecastBench53.2—
WinoGrande74.2%—

Math Gemma 4 31B IT leads

Claude 3 Haiku: 9.2 (#313), Gemma 4 31B IT: 41.3 (#72)

BenchmarkClaude 3 HaikuGemma 4 31B IT
OTIS Mock AIME 2024-20251.8%73.3%
LMArena Math11881464
MATH Level 514.9%—

Knowledge Gemma 4 31B IT leads

Claude 3 Haiku: 17.3 (#280), Gemma 4 31B IT: 37.9 (#141)

BenchmarkClaude 3 HaikuGemma 4 31B IT
GPQA Diamond36.3%75.8%
LMArena Expert11481465
SimpleQA Verified—10.4%
Confabulations34.2%—
Vectara Hallucination Rate—7.4%
MMLU73.8%—

Multimodal Gemma 4 31B IT leads

Claude 3 Haiku: 23.1 (#123), Gemma 4 31B IT: 41.0 (#32)

BenchmarkClaude 3 HaikuGemma 4 31B IT
LMArena Vision9501278
LMArena Document—1425
ScienceQA72%—

Multilingual Gemma 4 31B IT leads

Claude 3 Haiku: 36.1 (#238), Gemma 4 31B IT: 54.0 (#52)

BenchmarkClaude 3 HaikuGemma 4 31B IT
LMArena Non-English11781431
LMArena Chinese11551478
LMArena French11941436
LMArena Russian12041462
LMArena Spanish11651443
LMArena German1174—
LMArena Japanese1102—
LMArena Korean1108—

Instruction Following Gemma 4 31B IT leads

Claude 3 Haiku: 61.5 (#242), Gemma 4 31B IT: 75.6 (#58)

BenchmarkClaude 3 HaikuGemma 4 31B IT
LMArena Instruction Following11741433

Long Context Gemma 4 31B IT leads

Claude 3 Haiku: 36.1 (#232), Gemma 4 31B IT: 44.2 (#67)

BenchmarkClaude 3 HaikuGemma 4 31B IT
LMArena Longer Query11911446

Writing & Preference Gemma 4 31B IT leads

Claude 3 Haiku: 29.8 (#286), Gemma 4 31B IT: 60.7 (#92)

BenchmarkClaude 3 HaikuGemma 4 31B IT
LMArena Text11951443
LMArena Creative Writing11571416
EQ-Bench Creative Writing7171368
LMArena Multi-Turn11901452
EQ-Bench 4—1120

Frequently asked questions

Is Claude 3 Haiku better than Gemma 4 31B IT?

Gemma 4 31B IT is the stronger model overall, scoring 43.2 to 25.7 on the Noometry Index.

Is Claude 3 Haiku or Gemma 4 31B IT better for coding?

Gemma 4 31B IT scores higher on coding benchmarks: 41.9 versus 26.1 in the Noometry coding category.

How many benchmarks do Claude 3 Haiku and Gemma 4 31B IT share?

23 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Gemma 4 31B IT has 35.

Related comparisons

Go deeper