Noometry

Model comparison

Claude 3 Sonnet vs GPT-5.4 mini

GPT-5.4 mini is the stronger model overall, scoring 45.0 to 29.0 on the Noometry Index.

Last verified . 23 shared benchmarks.

Claude 3 Sonnet Anthropic

29.0

Rank #319 Confirmed

GPT-5.4 mini OpenAI

45.0

Rank #76 Confirmed

Summary

  • They share 23 benchmarks with published results for both. Claude 3 Sonnet scores higher in 0 categories and GPT-5.4 mini in 9 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in math, where GPT-5.4 mini leads 45.5 to 10.7.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 2.5% for Claude 3 Sonnet and 88.9% for GPT-5.4 mini.

Side by side

Claude 3 SonnetGPT-5.4 mini
ProviderAnthropicOpenAI
Noometry Index29.045.0
Released2024-02-292026-03-17
WeightsProprietaryProprietary
Context window—400K
Max output—128K
Input $ / M tokens—$0.75
Output $ / M tokens—$4.50
Results tracked3046

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

Category by category

Coding GPT-5.4 mini leads

Claude 3 Sonnet: 29.6 (#302), GPT-5.4 mini: 45.2 (#72)

BenchmarkClaude 3 SonnetGPT-5.4 mini
WeirdML10.2%60.3%
LMArena Coding12231438
FrontierCode—27%
LMArena WebDev—1397
SciCode—49.9%
BigCodeBench Instruct42.7%—
BigCodeBench Complete53.8%—
ALE-Bench—1,189
HumanEval+64%—
MBPP+69.3%—

Agentic & Tool Use Not comparable

Claude 3 Sonnet: —, GPT-5.4 mini: 29.9 (#81)

BenchmarkClaude 3 SonnetGPT-5.4 mini
DeepResearch Bench—36.3%

Reasoning GPT-5.4 mini leads

Claude 3 Sonnet: 20.5 (#237), GPT-5.4 mini: 30.4 (#85)

BenchmarkClaude 3 SonnetGPT-5.4 mini
LMArena Hard Prompts11971424
DTBench53.6%80%
Epoch Capabilities Index120.7148.84
ARC-AGI-2—18.9%
Kagi LLM Benchmark—37.9%
NYT Connections (extended)—61.8%
ARC-AGI-1—63.7%
CritPt—10%
Chess Puzzles—24%
Thematic Generalization—61.7%
Mystery Game Puzzles—11%
LMCA—40.8%
ForecastBench—57
WinoGrande75.1%—

Math GPT-5.4 mini leads

Claude 3 Sonnet: 10.7 (#310), GPT-5.4 mini: 45.5 (#75)

BenchmarkClaude 3 SonnetGPT-5.4 mini
OTIS Mock AIME 2024-20252.5%88.9%
LMArena Math12131419
FrontierMath (Tiers 1-3)—51.2%
FrontierMath Tier 4—9.8%
ProofBench—21%
MATH Level 518.2%—
FrontierMath (Feb 2025 set)—28.3%
FrontierMath Tier 4 (v1)—2.1%

Knowledge GPT-5.4 mini leads

Claude 3 Sonnet: 21.1 (#276), GPT-5.4 mini: 51.5 (#67)

BenchmarkClaude 3 SonnetGPT-5.4 mini
GPQA Diamond40.6%86.9%
LMArena Expert11731435
SimpleQA Verified—29.4%
Vectara Hallucination Rate—5.5%
MMLU75.9%—

Multimodal GPT-5.4 mini leads

Claude 3 Sonnet: 25.2 (#125), GPT-5.4 mini: 39.7 (#56)

BenchmarkClaude 3 SonnetGPT-5.4 mini
LMArena Vision9841245

Multilingual GPT-5.4 mini leads

Claude 3 Sonnet: 37.8 (#234), GPT-5.4 mini: 51.9 (#96)

BenchmarkClaude 3 SonnetGPT-5.4 mini
LMArena Non-English12051405
LMArena Chinese11891446
LMArena French12291440
LMArena German12041409
LMArena Japanese11311374
LMArena Korean11281368
LMArena Russian12271417
LMArena Spanish12041405

Instruction Following GPT-5.4 mini leads

Claude 3 Sonnet: 62.8 (#235), GPT-5.4 mini: 74.1 (#102)

BenchmarkClaude 3 SonnetGPT-5.4 mini
LMArena Instruction Following11991405

Long Context GPT-5.4 mini leads

Claude 3 Sonnet: 36.7 (#228), GPT-5.4 mini: 43.0 (#112)

BenchmarkClaude 3 SonnetGPT-5.4 mini
LMArena Longer Query12111407

Writing & Preference GPT-5.4 mini leads

Claude 3 Sonnet: 42.1 (#238), GPT-5.4 mini: 64.0 (#58)

BenchmarkClaude 3 SonnetGPT-5.4 mini
LMArena Text12181412
LMArena Creative Writing11861370
LMArena Multi-Turn12271429
EQ-Bench Creative Writing—1665

Frequently asked questions

Is Claude 3 Sonnet better than GPT-5.4 mini?

GPT-5.4 mini is the stronger model overall, scoring 45.0 to 29.0 on the Noometry Index.

Is Claude 3 Sonnet or GPT-5.4 mini better for coding?

GPT-5.4 mini scores higher on coding benchmarks: 45.2 versus 29.6 in the Noometry coding category.

How many benchmarks do Claude 3 Sonnet and GPT-5.4 mini share?

23 benchmarks have published results for both models. Claude 3 Sonnet has 30 scored results on Noometry and GPT-5.4 mini has 46.

Related comparisons

Go deeper