Noometry

Model comparison

Claude Opus 5 vs GPT-4

Claude Opus 5 is the stronger model overall, scoring 67.8 to 29.1 on the Noometry Index.

Last verified . 26 shared benchmarks.

Claude Opus 5 Anthropic

67.8

Rank #4 Confirmed

GPT-4 OpenAI

29.1

Rank #316 Confirmed

Summary

  • They share 26 benchmarks with published results for both. Claude Opus 5 scores higher in 8 categories and GPT-4 in 0 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where Claude Opus 5 leads 86.2 to 10.8.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 98.9% for Claude Opus 5 and 1.1% for GPT-4.
  • Claude Opus 5 is cheaper at $5 / $25 per million input/output tokens, against $30 / $60 for GPT-4.
  • Claude Opus 5 accepts more context: 1M tokens versus 8K.

Side by side

Claude Opus 5GPT-4
ProviderAnthropicOpenAI
Noometry Index67.829.1
Released2026-07-242023-03-14
WeightsProprietaryProprietary
Context window1M8K
Max output128K8K
Input $ / M tokens$5$30
Output $ / M tokens$25$60
Results tracked5738

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

Category by category

Coding Claude Opus 5 leads

Claude Opus 5: 67.5 (#5), GPT-4: 31.6 (#283)

BenchmarkClaude Opus 5GPT-4
WeirdML91.8%12.4%
LMArena Coding15341254
DeepSWE73.6%—
FrontierCode53.4%—
CursorBench46.6%—
LMArena WebDev1691—
FrontierSWE52%—
SciCode56.4%—
BigCodeBench Instruct—46%
BigCodeBench Complete—57.2%
ALE-Bench2,165—
HumanEval+—79.3%

Agentic & Tool Use Not comparable

Claude Opus 5: 55.6 (#1), GPT-4: —

BenchmarkClaude Opus 5GPT-4
APEX-Agents65.8%—
OSWorld 2.031.4%—
τ²-bench Banking48.7%—
PostTrainBench35%—
BALROG63.4%—
GBAEval79.6%—
GDP.pdf24%—
METR Time Horizons—36.1%
Vending-Bench 211,182—

Reasoning Claude Opus 5 leads

Claude Opus 5: 77.2 (#4), GPT-4: 17.8 (#289)

BenchmarkClaude Opus 5GPT-4
Chess Puzzles42%4%
LMArena Hard Prompts15261241
Mystery Game Puzzles59%12%
DTBench97.9%62.7%
LMCA64.5%17.1%
Epoch Capabilities Index162.78125.89
ARC-AGI-290.4%—
SimpleBench80.6%—
NYT Connections (extended)94.3%—
ARC-AGI-197.5%—
CritPt29.1%—
EBR-Bench45.7%—
Bench to the Future 30.12—
BIG-Bench Hard—75.1%
ForecastBench—57.8
HellaSwag—95.3%
WinoGrande—87.5%

Math Claude Opus 5 leads

Claude Opus 5: 86.2 (#8), GPT-4: 10.8 (#309)

BenchmarkClaude Opus 5GPT-4
OTIS Mock AIME 2024-202598.9%1.1%
LMArena Math15311269
FrontierMath (Tiers 1-3)85.6%—
FrontierMath Tier 473.2%—
ProofBench99%—
MATH Level 5—23%
GSM8K—92%

Knowledge Claude Opus 5 leads

Claude Opus 5: 66.8 (#9), GPT-4: 18.4 (#282)

BenchmarkClaude Opus 5GPT-4
GPQA Diamond93.9%35.7%
LMArena Expert15571211
SimpleQA Verified59.9%—
MMLU—86.4%
TriviaQA—84.8%

Multimodal Not comparable

Claude Opus 5: 50.8 (#8), GPT-4: —

BenchmarkClaude Opus 5GPT-4
LMArena Vision1319—
Blueprint-Bench 230.4%—
Furniture Assembly60.8%—
LMArena Document1516—

Multilingual Claude Opus 5 leads

Claude Opus 5: 58.8 (#4), GPT-4: 40.6 (#215)

BenchmarkClaude Opus 5GPT-4
LMArena Non-English15011246
LMArena Chinese15741242
LMArena French15191283
LMArena German15241251
LMArena Japanese15161209
LMArena Korean15211184
LMArena Russian15071251
LMArena Spanish15191261

Instruction Following Claude Opus 5 leads

Claude Opus 5: 79.2 (#7), GPT-4: 65.3 (#222)

BenchmarkClaude Opus 5GPT-4
LMArena Instruction Following15171241

Long Context Claude Opus 5 leads

Claude Opus 5: 46.5 (#21), GPT-4: 37.7 (#212)

BenchmarkClaude Opus 5GPT-4
LMArena Longer Query15151244

Writing & Preference Claude Opus 5 leads

Claude Opus 5: 79.2 (#1), GPT-4: 34.9 (#268)

BenchmarkClaude Opus 5GPT-4
LMArena Text15071263
LMArena Creative Writing14911244
EQ-Bench Creative Writing2133752
LMArena Multi-Turn14991257
EQ-Bench 41385—

Frequently asked questions

Is Claude Opus 5 better than GPT-4?

Claude Opus 5 is the stronger model overall, scoring 67.8 to 29.1 on the Noometry Index.

Which is cheaper, Claude Opus 5 or GPT-4?

Claude Opus 5 is cheaper. It lists at $5 per million input tokens and $25 per million output tokens; GPT-4 lists at $30 and $60.

Is Claude Opus 5 or GPT-4 better for coding?

Claude Opus 5 scores higher on coding benchmarks: 67.5 versus 31.6 in the Noometry coding category.

Which has the bigger context window?

Claude Opus 5 does, with 1M tokens against 8K.

How many benchmarks do Claude Opus 5 and GPT-4 share?

26 benchmarks have published results for both models. Claude Opus 5 has 57 scored results on Noometry and GPT-4 has 38.

Related comparisons

Go deeper