Noometry

Model comparison

Claude Opus 5 vs Claude Sonnet 4

Claude Opus 5 is the stronger model overall, scoring 67.8 to 40.8 on the Noometry Index. Claude Sonnet 4 costs 1.7× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.

Last verified . 31 shared benchmarks.

Claude Opus 5 Anthropic

67.8

Rank #4 Confirmed

Claude Sonnet 4 Anthropic

40.8

Rank #145 Confirmed

Summary

  • They share 31 benchmarks with published results for both. Claude Opus 5 scores higher in 10 categories and Claude Sonnet 4 in 0 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Claude Opus 5 leads 77.2 to 22.9.
  • The biggest single-benchmark swing is ARC-AGI-2: 90.4% for Claude Opus 5 and 5.9% for Claude Sonnet 4.
  • Claude Sonnet 4 is cheaper at $3 / $15 per million input/output tokens, against $5 / $25 for Claude Opus 5.
  • Claude Opus 5 accepts more context: 1M tokens versus 200K.

Side by side

Claude Opus 5Claude Sonnet 4
ProviderAnthropicAnthropic
Noometry Index67.840.8
Released2026-07-242025-05-22
WeightsProprietaryProprietary
Context window1M200K
Max output128K64K
Input $ / M tokens$5$3
Output $ / M tokens$25$15
Results tracked5758

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), Claude Sonnet 4: 43.5 (#88)

BenchmarkClaude Opus 5Claude Sonnet 4
SciCode56.4%40%
WeirdML91.8%46.1%
LMArena Coding15341414
ALE-Bench2,165655.35
DeepSWE73.6%—
FrontierCode53.4%—
SWE-bench Verified (bash only)—64.9%
Aider Polyglot—61.3%
CursorBench46.6%—
LMArena WebDev1691—
FrontierSWE52%—
GSO—4.9%

Agentic & Tool Use Claude Opus 5 leads

Claude Opus 5: 55.6 (#1), Claude Sonnet 4: 38.5 (#31)

BenchmarkClaude Opus 5Claude Sonnet 4
APEX-Agents65.8%—
OSWorld 2.031.4%—
TheAgentCompany—33.1%
τ²-bench Banking48.7%—
Cybench—35%
DeepResearch Bench—46.6%
OSWorld—43.9%
PostTrainBench35%—
BALROG63.4%—
GBAEval79.6%—
GDP.pdf24%—
METR Time Horizons—62%
Vending-Bench 211,182—

Reasoning Claude Opus 5 leads

Claude Opus 5: 77.2 (#4), Claude Sonnet 4: 22.9 (#187)

BenchmarkClaude Opus 5Claude Sonnet 4
ARC-AGI-290.4%5.9%
SimpleBench80.6%45.5%
ARC-AGI-197.5%40%
CritPt29.1%0.3%
LMArena Hard Prompts15261372
DTBench97.9%77.1%
LMCA64.5%29%
Epoch Capabilities Index162.78141.69
Kagi LLM Benchmark—73%
NYT Connections (extended)94.3%—
Chess Puzzles42%—
EnigmaEval—3.1%
EBR-Bench45.7%—
Mystery Game Puzzles59%—
Bench to the Future 30.12—
ForecastBench—60.2

Math Claude Opus 5 leads

Claude Opus 5: 86.2 (#8), Claude Sonnet 4: 43.3 (#80)

BenchmarkClaude Opus 5Claude Sonnet 4
OTIS Mock AIME 2024-202598.9%71.1%
LMArena Math15311375
FrontierMath (Tiers 1-3)85.6%—
FrontierMath Tier 473.2%—
ProofBench99%—
Omni-MATH—60.2%
MATH Level 5—84.4%
FrontierMath (Feb 2025 set)—4.1%
FrontierMath Tier 4 (v1)—0%

Knowledge Claude Opus 5 leads

Claude Opus 5: 66.8 (#9), Claude Sonnet 4: 41.8 (#108)

BenchmarkClaude Opus 5Claude Sonnet 4
GPQA Diamond93.9%79.2%
LMArena Expert15571372
Humanity's Last Exam—7.8%
SimpleQA Verified59.9%—
MMLU-Pro—84.3%
Confabulations—13.2%
Vectara Hallucination Rate—10.3%
GPQA (HELM)—70.6%

Multimodal Claude Opus 5 leads

Claude Opus 5: 50.8 (#8), Claude Sonnet 4: 26.2 (#121)

BenchmarkClaude Opus 5Claude Sonnet 4
LMArena Vision13191191
GeoBench—37%
VPCT—34%
Blueprint-Bench 230.4%—
Furniture Assembly60.8%—
LMArena Document1516—
MindCube—44.8%

Multilingual Claude Opus 5 leads

Claude Opus 5: 58.8 (#4), Claude Sonnet 4: 46.7 (#156)

BenchmarkClaude Opus 5Claude Sonnet 4
LMArena Non-English15011333
LMArena Chinese15741350
LMArena French15191363
LMArena German15241331
LMArena Japanese15161302
LMArena Korean15211291
LMArena Russian15071355
LMArena Spanish15191357

Instruction Following Claude Opus 5 leads

Claude Opus 5: 79.2 (#7), Claude Sonnet 4: 71.7 (#145)

BenchmarkClaude Opus 5Claude Sonnet 4
LMArena Instruction Following15171376
IFEval—84%

Long Context Claude Opus 5 leads

Claude Opus 5: 46.5 (#21), Claude Sonnet 4: 33.7 (#259)

BenchmarkClaude Opus 5Claude Sonnet 4
LMArena Longer Query15151398
Fiction.LiveBench—46.9%

Writing & Preference Claude Opus 5 leads

Claude Opus 5: 79.2 (#1), Claude Sonnet 4: 57.1 (#132)

BenchmarkClaude Opus 5Claude Sonnet 4
LMArena Text15071351
LMArena Creative Writing14911345
EQ-Bench Creative Writing21331483
LMArena Multi-Turn14991376
Short-Story Creative Writing—81.4%
WildBench—83.8%
EQ-Bench 41385—

Frequently asked questions

Is Claude Opus 5 better than Claude Sonnet 4?

Claude Opus 5 is the stronger model overall, scoring 67.8 to 40.8 on the Noometry Index. Claude Sonnet 4 costs 1.7× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.

Which is cheaper, Claude Opus 5 or Claude Sonnet 4?

Claude Sonnet 4 is cheaper. It lists at $3 per million input tokens and $15 per million output tokens; Claude Opus 5 lists at $5 and $25.

Is Claude Opus 5 or Claude Sonnet 4 better for coding?

Claude Opus 5 scores higher on coding benchmarks: 67.5 versus 43.5 in the Noometry coding category.

Which has the bigger context window?

Claude Opus 5 does, with 1M tokens against 200K.

How many benchmarks do Claude Opus 5 and Claude Sonnet 4 share?

31 benchmarks have published results for both models. Claude Opus 5 has 57 scored results on Noometry and Claude Sonnet 4 has 58.

Related comparisons

Go deeper