Noometry

Model comparison

Claude 3 Haiku vs Grok 4.5

Grok 4.5 is the stronger model overall, scoring 55.0 to 25.9 on the Noometry Index.

Last verified . 26 shared benchmarks.

Claude 3 Haiku Anthropic

25.9

Rank #340 Confirmed

Grok 4.5 xAI

55.0

Rank #25 Confirmed

Summary

  • They share 26 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and Grok 4.5 in 9 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in math, where Grok 4.5 leads 60.9 to 9.8.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.8% for Claude 3 Haiku and 97.8% for Grok 4.5.

Side by side

Claude 3 HaikuGrok 4.5
ProviderAnthropicxAI
Noometry Index25.955.0
Released2024-03-072026-07-08
WeightsProprietaryProprietary
Context window—500K
Max output—500K
Input $ / M tokens—$2
Output $ / M tokens—$6
Results tracked3752

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

Category by category

Coding Grok 4.5 leads

Claude 3 Haiku: 26.4 (#325), Grok 4.5: 52.2 (#35)

BenchmarkClaude 3 HaikuGrok 4.5
WeirdML9.8%46.4%
LMArena Coding11991474
DeepSWE—53.8%
FrontierCode—42.4%
LMArena WebDev—1553
SciCode—54.1%
BigCodeBench Instruct39.4%—
BigCodeBench Complete50.1%—
CadEval12%—
ALE-Bench—1,309
HumanEval+68.9%—
MBPP+68.8%—

Agentic & Tool Use Not comparable

Claude 3 Haiku: —, Grok 4.5: 44.4 (#17)

BenchmarkClaude 3 HaikuGrok 4.5
APEX-Agents—56.2%
τ²-bench Banking—47.9%
PostTrainBench—23.4%
GBAEval—65.4%
GDP.pdf—14%
LMArena Search—1213
Vending-Bench 2—3,887

Reasoning Grok 4.5 leads

Claude 3 Haiku: 16.3 (#307), Grok 4.5: 56.1 (#25)

BenchmarkClaude 3 HaikuGrok 4.5
Kagi LLM Benchmark34.2%83.5%
LMArena Hard Prompts11741462
DTBench50.1%96.5%
LMCA8.8%45.2%
Epoch Capabilities Index118.35153.92
ARC-AGI-2—52.6%
SimpleBench—70%
NYT Connections (extended)—79.9%
ARC-AGI-1—87.2%
CritPt—15.4%
Chess Puzzles—36%
Surface Evolver Bench—74.4%
ForecastBench53.2—
WinoGrande74.2%—

Math Grok 4.5 leads

Claude 3 Haiku: 9.8 (#319), Grok 4.5: 60.9 (#35)

BenchmarkClaude 3 HaikuGrok 4.5
OTIS Mock AIME 2024-20251.8%97.8%
LMArena Math11881459
FrontierMath (Tiers 1-3)—57.2%
FrontierMath Tier 4—24.4%
ProofBench—31%
MATH Level 514.9%—

Knowledge Grok 4.5 leads

Claude 3 Haiku: 17.3 (#285), Grok 4.5: 62.3 (#24)

BenchmarkClaude 3 HaikuGrok 4.5
GPQA Diamond36.3%93.4%
LMArena Expert11481466
SimpleQA Verified—48.3%
Confabulations34.2%—
MMLU73.8%—

Multimodal Grok 4.5 leads

Claude 3 Haiku: 23.6 (#128), Grok 4.5: 37.6 (#72)

BenchmarkClaude 3 HaikuGrok 4.5
LMArena Vision9501288
Blueprint-Bench 2—27.3%
Furniture Assembly—22.5%
LMArena Document—1452
ScienceQA72%—

Multilingual Grok 4.5 leads

Claude 3 Haiku: 36.0 (#243), Grok 4.5: 54.4 (#42)

BenchmarkClaude 3 HaikuGrok 4.5
LMArena Non-English11781440
LMArena Chinese11551496
LMArena French11951456
LMArena German11741446
LMArena Japanese11021428
LMArena Korean11091404
LMArena Russian12041448
LMArena Spanish11661450

Instruction Following Grok 4.5 leads

Claude 3 Haiku: 61.3 (#247), Grok 4.5: 76.0 (#48)

BenchmarkClaude 3 HaikuGrok 4.5
LMArena Instruction Following11731446

Long Context Grok 4.5 leads

Claude 3 Haiku: 36.1 (#237), Grok 4.5: 44.8 (#56)

BenchmarkClaude 3 HaikuGrok 4.5
LMArena Longer Query11901463

Writing & Preference Grok 4.5 leads

Claude 3 Haiku: 29.7 (#291), Grok 4.5: 65.8 (#42)

BenchmarkClaude 3 HaikuGrok 4.5
LMArena Text11951448
LMArena Creative Writing11571442
EQ-Bench Creative Writing7171579
LMArena Multi-Turn11901456

Frequently asked questions

Is Claude 3 Haiku better than Grok 4.5?

Grok 4.5 is the stronger model overall, scoring 55.0 to 25.9 on the Noometry Index.

Is Claude 3 Haiku or Grok 4.5 better for coding?

Grok 4.5 scores higher on coding benchmarks: 52.2 versus 26.4 in the Noometry coding category.

How many benchmarks do Claude 3 Haiku and Grok 4.5 share?

26 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Grok 4.5 has 52.

Related comparisons

Go deeper