Noometry

Model comparison

Claude Opus 4.7 vs Claude Sonnet 4.5

Claude Opus 4.7 is the stronger model overall, scoring 58.3 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 4.7's lead doesn't matter for your workload.

Last verified . 52 shared benchmarks.

Claude Opus 4.7 Anthropic

58.3

Rank #19 Confirmed

Claude Sonnet 4.5 Anthropic

44.1

Rank #81 Confirmed

Summary

  • They share 52 benchmarks with published results for both. Claude Opus 4.7 scores higher in 10 categories and Claude Sonnet 4.5 in 0 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in math, where Claude Opus 4.7 leads 66.7 to 32.3.
  • The biggest single-benchmark swing is ARC-AGI-2: 75.8% for Claude Opus 4.7 and 13.6% 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 4.7.
  • Claude Opus 4.7 accepts more context: 1M tokens versus 200K.

Side by side

Claude Opus 4.7Claude Sonnet 4.5
ProviderAnthropicAnthropic
Noometry Index58.344.1
Released2026-04-142025-09-29
WeightsProprietaryProprietary
Context window1M200K
Max output128K64K
Input $ / M tokens$5$3
Output $ / M tokens$25$15
Results tracked6673

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

Category by category

Coding Claude Opus 4.7 leads

Claude Opus 4.7: 59.6 (#13), Claude Sonnet 4.5: 47.3 (#61)

BenchmarkClaude Opus 4.7Claude Sonnet 4.5
SWE-bench Verified83.5%71.3%
LMArena WebDev15581393
SciCode54.5%44.7%
GSO44.1%14.7%
WeirdML76.4%47.7%
LMArena Coding15181489
ALE-Bench1,323796.15
FrontierCode38.5%—
SWE-bench Verified (bash only)—71.4%
SWE-bench Multilingual—67%
MirrorCode31.1%—
AlgoTune—1.52

Agentic & Tool Use Claude Opus 4.7 leads

Claude Opus 4.7: 47.9 (#10), Claude Sonnet 4.5: 38.3 (#32)

BenchmarkClaude Opus 4.7Claude Sonnet 4.5
Terminal-Bench80.2%46.5%
τ²-bench Banking40.2%25.3%
LMArena Search12331159
Vending-Bench 210,9373,839
APEX-Agents49.2%—
Berkeley Function Calling Leaderboard—73.2%
OSWorld 2.018.2%—
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%
PostTrainBench28.6%—
ExploitBench26.5%—
GBAEval43.8%—
GDP.pdf21%—
METR Time Horizons—67.4%

Reasoning Claude Opus 4.7 leads

Claude Opus 4.7: 53.8 (#29), Claude Sonnet 4.5: 26.9 (#125)

BenchmarkClaude Opus 4.7Claude Sonnet 4.5
ARC-AGI-275.8%13.6%
SimpleBench61.7%54.3%
Kagi LLM Benchmark80.7%57.9%
NYT Connections (extended)39%37.3%
ARC-AGI-193.5%63.7%
CritPt12%1.1%
Chess Puzzles30%12%
EBR-Bench19%2.4%
LMArena Hard Prompts15061462
Mystery Game Puzzles28%17%
DTBench94.7%83.2%
LMCA52.2%38.8%
Epoch Capabilities Index156.25146.84
ForecastBench60.361.9
EnigmaEval—6%
Thematic Generalization72.8%—

Math Claude Opus 4.7 leads

Claude Opus 4.7: 66.7 (#26), Claude Sonnet 4.5: 32.3 (#216)

BenchmarkClaude Opus 4.7Claude Sonnet 4.5
FrontierMath (Tiers 1-3)70.2%23.9%
FrontierMath Tier 431.7%2.4%
OTIS Mock AIME 2024-202597.8%77.8%
ProofBench54%19%
LMArena Math14991449
FrontierMath (Feb 2025 set)43.8%15.2%
FrontierMath Tier 4 (v1)22.9%4.2%
MathArena Final-Answer Competitions73.6%—
Omni-MATH—55.3%
MATH Level 5—97.7%

Knowledge Claude Opus 4.7 leads

Claude Opus 4.7: 62.6 (#23), Claude Sonnet 4.5: 48.4 (#76)

BenchmarkClaude Opus 4.7Claude Sonnet 4.5
GPQA Diamond90.2%82.3%
Humanity's Last Exam36.2%13.7%
SimpleQA Verified51.7%30.7%
Vectara Hallucination Rate12%12%
LMArena Expert15211482
MMLU-Pro—86.9%
GPQA (HELM)—68.6%

Multimodal Claude Opus 4.7 leads

Claude Opus 4.7: 41.2 (#38), Claude Sonnet 4.5: 34.8 (#89)

BenchmarkClaude Opus 4.7Claude Sonnet 4.5
LMArena Document14951450
LMArena Vision1316—
VPCT—39.8%
Blueprint-Bench 224.5%—
Furniture Assembly33.3%—

Multilingual Claude Opus 4.7 leads

Claude Opus 4.7: 57.3 (#10), Claude Sonnet 4.5: 53.4 (#69)

BenchmarkClaude Opus 4.7Claude Sonnet 4.5
LMArena Non-English14801425
LMArena Chinese15311459
LMArena French15031458
LMArena German14951427
LMArena Japanese14721390
LMArena Korean14641403
LMArena Russian14941437
LMArena Spanish14951457

Instruction Following Claude Opus 4.7 leads

Claude Opus 4.7: 78.4 (#10), Claude Sonnet 4.5: 75.0 (#78)

BenchmarkClaude Opus 4.7Claude Sonnet 4.5
LMArena Instruction Following14981459
IFEval—85%

Long Context Too close to call

Claude Opus 4.7: 46.2 (#25), Claude Sonnet 4.5: 45.2 (#46)

BenchmarkClaude Opus 4.7Claude Sonnet 4.5
LMArena Longer Query15051476

Writing & Preference Claude Opus 4.7 leads

Claude Opus 4.7: 75.1 (#8), Claude Sonnet 4.5: 66.5 (#34)

BenchmarkClaude Opus 4.7Claude Sonnet 4.5
LMArena Text14901439
LMArena Creative Writing14861442
EQ-Bench Creative Writing19141678
LMArena Multi-Turn15051465
WildBench—85.4%
EQ-Bench 41311—

Frequently asked questions

Is Claude Opus 4.7 better than Claude Sonnet 4.5?

Claude Opus 4.7 is the stronger model overall, scoring 58.3 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 4.7's lead doesn't matter for your workload.

Which is cheaper, Claude Opus 4.7 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 4.7 lists at $5 and $25.

Is Claude Opus 4.7 or Claude Sonnet 4.5 better for coding?

Claude Opus 4.7 scores higher on coding benchmarks: 59.6 versus 47.3 in the Noometry coding category.

Which has the bigger context window?

Claude Opus 4.7 does, with 1M tokens against 200K.

How many benchmarks do Claude Opus 4.7 and Claude Sonnet 4.5 share?

52 benchmarks have published results for both models. Claude Opus 4.7 has 66 scored results on Noometry and Claude Sonnet 4.5 has 73.

Related comparisons

Go deeper