Noometry

Model comparison

Claude Opus 4.6 vs Claude Sonnet 4

Claude Opus 4.6 is the stronger model overall, scoring 58.2 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 4.6's lead doesn't matter for your workload.

Last verified . 41 shared benchmarks.

Claude Opus 4.6 Anthropic

58.2

Rank #20 Confirmed

Claude Sonnet 4 Anthropic

40.8

Rank #145 Confirmed

Summary

  • They share 41 benchmarks with published results for both. Claude Opus 4.6 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 4.6 leads 57.8 to 22.9.
  • The biggest single-benchmark swing is ARC-AGI-2: 69.2% for Claude Opus 4.6 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 4.6.
  • Claude Opus 4.6 accepts more context: 1M tokens versus 200K.

Side by side

Claude Opus 4.6Claude Sonnet 4
ProviderAnthropicAnthropic
Noometry Index58.240.8
Released2026-02-042025-05-22
WeightsProprietaryProprietary
Context window1M200K
Max output128K64K
Input $ / M tokens$5$3
Output $ / M tokens$25$15
Results tracked6858

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

Category by category

Coding Claude Opus 4.6 leads

Claude Opus 4.6: 57.2 (#20), Claude Sonnet 4: 43.5 (#88)

BenchmarkClaude Opus 4.6Claude Sonnet 4
SWE-bench Verified (bash only)75.6%64.9%
GSO41.2%4.9%
WeirdML78%46.1%
LMArena Coding15361414
ALE-Bench996.5655.35
SWE-bench Verified78.7%—
FrontierCode26.6%—
Aider Polyglot—61.3%
LMArena WebDev1547—
SWE-bench Multilingual72%—
SciCode—40%
AlgoTune1.47—

Agentic & Tool Use Claude Opus 4.6 leads

Claude Opus 4.6: 51.1 (#4), Claude Sonnet 4: 38.5 (#31)

BenchmarkClaude Opus 4.6Claude Sonnet 4
Cybench93%35%
DeepResearch Bench55.3%46.6%
METR Time Horizons78.9%62%
Terminal-Bench79.8%—
APEX-Agents46.3%—
Remote Labor Index4.2%—
TheAgentCompany—33.1%
τ²-bench Banking27.3%—
OSWorld—43.9%
GBAEval44.1%—
LMArena Search1253—
Vending-Bench 28,018—

Reasoning Claude Opus 4.6 leads

Claude Opus 4.6: 57.8 (#23), Claude Sonnet 4: 22.9 (#187)

BenchmarkClaude Opus 4.6Claude Sonnet 4
ARC-AGI-269.2%5.9%
SimpleBench67.6%45.5%
Kagi LLM Benchmark83.6%73%
ARC-AGI-194%40%
EnigmaEval7.6%3.1%
LMArena Hard Prompts15271372
DTBench91.2%77.1%
LMCA55.8%29%
Epoch Capabilities Index155.24141.69
ForecastBench6060.2
NYT Connections (extended)92.1%—
CritPt—0.3%
Chess Puzzles17%—
Thematic Generalization80.6%—
EBR-Bench12.7%—
Mystery Game Puzzles25%—

Math Claude Opus 4.6 leads

Claude Opus 4.6: 63.0 (#31), Claude Sonnet 4: 43.3 (#80)

BenchmarkClaude Opus 4.6Claude Sonnet 4
OTIS Mock AIME 2024-202594.4%71.1%
LMArena Math15191375
FrontierMath (Feb 2025 set)40.7%4.1%
FrontierMath Tier 4 (v1)22.9%0%
FrontierMath (Tiers 1-3)66%—
FrontierMath Tier 426.8%—
MathArena Final-Answer Competitions78.5%—
ProofBench50%—
Omni-MATH—60.2%
MATH Level 5—84.4%

Knowledge Claude Opus 4.6 leads

Claude Opus 4.6: 61.9 (#26), Claude Sonnet 4: 41.8 (#108)

BenchmarkClaude Opus 4.6Claude Sonnet 4
GPQA Diamond90.5%79.2%
Humanity's Last Exam34.4%7.8%
Vectara Hallucination Rate12.2%10.3%
LMArena Expert15461372
SimpleQA Verified47%—
MMLU-Pro—84.3%
Confabulations—13.2%
GPQA (HELM)—70.6%

Multimodal Claude Opus 4.6 leads

Claude Opus 4.6: 37.3 (#74), Claude Sonnet 4: 26.2 (#121)

BenchmarkClaude Opus 4.6Claude Sonnet 4
LMArena Vision13161191
GeoBench—37%
VPCT—34%
Furniture Assembly28.3%—
LMArena Document1507—
MindCube—44.8%

Multilingual Claude Opus 4.6 leads

Claude Opus 4.6: 57.9 (#6), Claude Sonnet 4: 46.7 (#156)

BenchmarkClaude Opus 4.6Claude Sonnet 4
LMArena Non-English14891333
LMArena Chinese15511350
LMArena French15131363
LMArena German15021331
LMArena Japanese14841302
LMArena Korean14641291
LMArena Russian14971355
LMArena Spanish15101357

Instruction Following Claude Opus 4.6 leads

Claude Opus 4.6: 79.5 (#4), Claude Sonnet 4: 71.7 (#145)

BenchmarkClaude Opus 4.6Claude Sonnet 4
LMArena Instruction Following15231376
IFEval—84%

Long Context Claude Opus 4.6 leads

Claude Opus 4.6: 48.1 (#13), Claude Sonnet 4: 33.7 (#259)

BenchmarkClaude Opus 4.6Claude Sonnet 4
LMArena Longer Query15201398
Fiction.LiveBench—46.9%
CL-bench20.7%—
CL-bench Life17%—

Writing & Preference Claude Opus 4.6 leads

Claude Opus 4.6: 73.5 (#10), Claude Sonnet 4: 57.1 (#132)

BenchmarkClaude Opus 4.6Claude Sonnet 4
LMArena Text15031351
LMArena Creative Writing15051345
EQ-Bench Creative Writing18091483
LMArena Multi-Turn15131376
Short-Story Creative Writing—81.4%
WildBench—83.8%
EQ-Bench 41223—

Frequently asked questions

Is Claude Opus 4.6 better than Claude Sonnet 4?

Claude Opus 4.6 is the stronger model overall, scoring 58.2 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 4.6's lead doesn't matter for your workload.

Which is cheaper, Claude Opus 4.6 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 4.6 lists at $5 and $25.

Is Claude Opus 4.6 or Claude Sonnet 4 better for coding?

Claude Opus 4.6 scores higher on coding benchmarks: 57.2 versus 43.5 in the Noometry coding category.

Which has the bigger context window?

Claude Opus 4.6 does, with 1M tokens against 200K.

How many benchmarks do Claude Opus 4.6 and Claude Sonnet 4 share?

41 benchmarks have published results for both models. Claude Opus 4.6 has 68 scored results on Noometry and Claude Sonnet 4 has 58.

Related comparisons

Go deeper