Noometry

Model comparison

Claude Opus 5 vs Claude Sonnet 4.5

Claude Opus 5 is the stronger model overall, scoring 67.8 to 44.1 on the Noometry Index. Claude Sonnet 4.5 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 . 42 shared benchmarks.

Claude Opus 5 Anthropic

67.8

Rank #4 Confirmed

Claude Sonnet 4.5 Anthropic

44.1

Rank #81 Confirmed

Summary

  • They share 42 benchmarks with published results for both. Claude Opus 5 scores higher in 10 categories and Claude Sonnet 4.5 in 0 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in math, where Claude Opus 5 leads 86.2 to 32.3.
  • The biggest single-benchmark swing is ProofBench: 99% for Claude Opus 5 and 19% for Claude Sonnet 4.5.
  • Claude Sonnet 4.5 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 4.5
ProviderAnthropicAnthropic
Noometry Index67.844.1
Released2026-07-242025-09-29
WeightsProprietaryProprietary
Context window1M200K
Max output128K64K
Input $ / M tokens$5$3
Output $ / M tokens$25$15
Results tracked5773

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 4.5: 47.3 (#61)

BenchmarkClaude Opus 5Claude Sonnet 4.5
LMArena WebDev16911393
SciCode56.4%44.7%
WeirdML91.8%47.7%
LMArena Coding15341489
ALE-Bench2,165796.15
SWE-bench Verified—71.3%
DeepSWE73.6%—
FrontierCode53.4%—
SWE-bench Verified (bash only)—71.4%
CursorBench46.6%—
SWE-bench Multilingual—67%
FrontierSWE52%—
GSO—14.7%
AlgoTune—1.52

Agentic & Tool Use Claude Opus 5 leads

Claude Opus 5: 55.6 (#1), Claude Sonnet 4.5: 38.3 (#32)

BenchmarkClaude Opus 5Claude Sonnet 4.5
τ²-bench Banking48.7%25.3%
Vending-Bench 211,1823,839
Terminal-Bench—46.5%
APEX-Agents65.8%—
Berkeley Function Calling Leaderboard—73.2%
OSWorld 2.031.4%—
GDPval—42.5%
Remote Labor Index—2.1%
τ²-bench Airline—72%
τ²-bench Retail—72.4%
τ²-bench Telecom—84.9%
Cybench—60%
DeepResearch Bench—52.6%
OSWorld—62.9%
PostTrainBench35%—
BALROG63.4%—
GBAEval79.6%—
GDP.pdf24%—
LMArena Search—1159
METR Time Horizons—67.4%

Reasoning Claude Opus 5 leads

Claude Opus 5: 77.2 (#4), Claude Sonnet 4.5: 26.9 (#125)

BenchmarkClaude Opus 5Claude Sonnet 4.5
ARC-AGI-290.4%13.6%
SimpleBench80.6%54.3%
NYT Connections (extended)94.3%37.3%
ARC-AGI-197.5%63.7%
CritPt29.1%1.1%
Chess Puzzles42%12%
EBR-Bench45.7%2.4%
LMArena Hard Prompts15261462
Mystery Game Puzzles59%17%
DTBench97.9%83.2%
LMCA64.5%38.8%
Epoch Capabilities Index162.78146.84
Kagi LLM Benchmark—57.9%
EnigmaEval—6%
Bench to the Future 30.12—
ForecastBench—61.9

Math Claude Opus 5 leads

Claude Opus 5: 86.2 (#8), Claude Sonnet 4.5: 32.3 (#216)

BenchmarkClaude Opus 5Claude Sonnet 4.5
FrontierMath (Tiers 1-3)85.6%23.9%
FrontierMath Tier 473.2%2.4%
OTIS Mock AIME 2024-202598.9%77.8%
ProofBench99%19%
LMArena Math15311449
Omni-MATH—55.3%
MATH Level 5—97.7%
FrontierMath (Feb 2025 set)—15.2%
FrontierMath Tier 4 (v1)—4.2%

Knowledge Claude Opus 5 leads

Claude Opus 5: 66.8 (#9), Claude Sonnet 4.5: 48.4 (#76)

BenchmarkClaude Opus 5Claude Sonnet 4.5
GPQA Diamond93.9%82.3%
SimpleQA Verified59.9%30.7%
LMArena Expert15571482
Humanity's Last Exam—13.7%
MMLU-Pro—86.9%
Vectara Hallucination Rate—12%
GPQA (HELM)—68.6%

Multimodal Claude Opus 5 leads

Claude Opus 5: 50.8 (#8), Claude Sonnet 4.5: 34.8 (#89)

BenchmarkClaude Opus 5Claude Sonnet 4.5
LMArena Document15161450
LMArena Vision1319—
VPCT—39.8%
Blueprint-Bench 230.4%—
Furniture Assembly60.8%—

Multilingual Claude Opus 5 leads

Claude Opus 5: 58.8 (#4), Claude Sonnet 4.5: 53.4 (#69)

BenchmarkClaude Opus 5Claude Sonnet 4.5
LMArena Non-English15011425
LMArena Chinese15741459
LMArena French15191458
LMArena German15241427
LMArena Japanese15161390
LMArena Korean15211403
LMArena Russian15071437
LMArena Spanish15191457

Instruction Following Claude Opus 5 leads

Claude Opus 5: 79.2 (#7), Claude Sonnet 4.5: 75.0 (#78)

BenchmarkClaude Opus 5Claude Sonnet 4.5
LMArena Instruction Following15171459
IFEval—85%

Long Context Claude Opus 5 leads

Claude Opus 5: 46.5 (#21), Claude Sonnet 4.5: 45.2 (#46)

BenchmarkClaude Opus 5Claude Sonnet 4.5
LMArena Longer Query15151476

Writing & Preference Claude Opus 5 leads

Claude Opus 5: 79.2 (#1), Claude Sonnet 4.5: 66.5 (#34)

BenchmarkClaude Opus 5Claude Sonnet 4.5
LMArena Text15071439
LMArena Creative Writing14911442
EQ-Bench Creative Writing21331678
LMArena Multi-Turn14991465
WildBench—85.4%
EQ-Bench 41385—

Frequently asked questions

Is Claude Opus 5 better than Claude Sonnet 4.5?

Claude Opus 5 is the stronger model overall, scoring 67.8 to 44.1 on the Noometry Index. Claude Sonnet 4.5 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 4.5?

Claude Sonnet 4.5 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 4.5 better for coding?

Claude Opus 5 scores higher on coding benchmarks: 67.5 versus 47.3 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 4.5 share?

42 benchmarks have published results for both models. Claude Opus 5 has 57 scored results on Noometry and Claude Sonnet 4.5 has 73.

Related comparisons

Go deeper