Noometry

Model comparison

Claude Sonnet 4.5 vs Claude Sonnet 5

Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 44.1 on the Noometry Index.

Last verified . 41 shared benchmarks.

Claude Sonnet 4.5 Anthropic

44.1

Rank #81 Confirmed

Claude Sonnet 5 Anthropic

54.6

Rank #29 Confirmed

Summary

  • They share 41 benchmarks with published results for both. Claude Sonnet 4.5 scores higher in 1 category and Claude Sonnet 5 in 9 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where Claude Sonnet 5 leads 66.2 to 32.3.
  • The biggest single-benchmark swing is ProofBench: 19% for Claude Sonnet 4.5 and 77% for Claude Sonnet 5.
  • Claude Sonnet 5 is cheaper at $2 / $10 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.5.
  • Claude Sonnet 5 accepts more context: 1M tokens versus 200K.

Side by side

Claude Sonnet 4.5Claude Sonnet 5
ProviderAnthropicAnthropic
Noometry Index44.154.6
Released2025-09-292026-06-29
WeightsProprietaryProprietary
Context window200K1M
Max output64K128K
Input $ / M tokens$3$2
Output $ / M tokens$15$10
Results tracked7351

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

Category by category

Coding Claude Sonnet 5 leads

Claude Sonnet 4.5: 47.3 (#61), Claude Sonnet 5: 55.5 (#26)

BenchmarkClaude Sonnet 4.5Claude Sonnet 5
LMArena WebDev13931541
SciCode44.7%54.3%
GSO14.7%37.3%
WeirdML47.7%68.8%
LMArena Coding14891483
ALE-Bench796.151,463
SWE-bench Verified71.3%—
DeepSWE—53.8%
FrontierCode—42.7%
SWE-bench Verified (bash only)71.4%—
CursorBench—34.1%
SWE-bench Multilingual67%—
AlgoTune1.52—

Agentic & Tool Use Claude Sonnet 5 leads

Claude Sonnet 4.5: 38.3 (#32), Claude Sonnet 5: 42.8 (#18)

BenchmarkClaude Sonnet 4.5Claude Sonnet 5
LMArena Search11591194
Vending-Bench 23,8396,378
Terminal-Bench46.5%—
APEX-Agents—54.5%
Berkeley Function Calling Leaderboard73.2%—
GDPval42.5%—
Remote Labor Index2.1%—
τ²-bench Airline72%—
τ²-bench Banking25.3%—
τ²-bench Retail72.4%—
τ²-bench Telecom84.9%—
Cybench60%—
DeepResearch Bench52.6%—
OSWorld62.9%—
GBAEval—65.3%
METR Time Horizons67.4%—

Reasoning Claude Sonnet 5 leads

Claude Sonnet 4.5: 26.9 (#125), Claude Sonnet 5: 49.1 (#39)

BenchmarkClaude Sonnet 4.5Claude Sonnet 5
SimpleBench54.3%60.6%
NYT Connections (extended)37.3%75.1%
CritPt1.1%16.9%
Chess Puzzles12%35%
LMArena Hard Prompts14621461
Mystery Game Puzzles17%35%
DTBench83.2%92.5%
LMCA38.8%50%
Epoch Capabilities Index146.84156.21
ForecastBench61.961.1
ARC-AGI-213.6%—
Kagi LLM Benchmark57.9%—
ARC-AGI-163.7%—
EnigmaEval6%—
EBR-Bench2.4%—
Surface Evolver Bench—60%
Bench to the Future 3—0.14

Math Claude Sonnet 5 leads

Claude Sonnet 4.5: 32.3 (#216), Claude Sonnet 5: 66.2 (#27)

BenchmarkClaude Sonnet 4.5Claude Sonnet 5
FrontierMath (Tiers 1-3)23.9%65.6%
FrontierMath Tier 42.4%29.3%
OTIS Mock AIME 2024-202577.8%94.7%
ProofBench19%77%
LMArena Math14491467
Omni-MATH55.3%—
MATH Level 597.7%—
FrontierMath (Feb 2025 set)15.2%—
FrontierMath Tier 4 (v1)4.2%—

Knowledge Claude Sonnet 5 leads

Claude Sonnet 4.5: 48.4 (#76), Claude Sonnet 5: 55.6 (#47)

BenchmarkClaude Sonnet 4.5Claude Sonnet 5
GPQA Diamond82.3%90.5%
SimpleQA Verified30.7%33.7%
LMArena Expert14821490
Humanity's Last Exam13.7%—
MMLU-Pro86.9%—
Vectara Hallucination Rate12%—
GPQA (HELM)68.6%—

Multimodal Claude Sonnet 5 leads

Claude Sonnet 4.5: 34.8 (#89), Claude Sonnet 5: 42.4 (#31)

BenchmarkClaude Sonnet 4.5Claude Sonnet 5
LMArena Document14501466
LMArena Vision—1274
VPCT39.8%—
Blueprint-Bench 2—24.9%

Multilingual Too close to call

Claude Sonnet 4.5: 53.4 (#69), Claude Sonnet 5: 53.8 (#55)

BenchmarkClaude Sonnet 4.5Claude Sonnet 5
LMArena Non-English14251431
LMArena Chinese14591477
LMArena French14581460
LMArena German14271440
LMArena Japanese13901422
LMArena Korean14031411
LMArena Russian14371451
LMArena Spanish14571437

Instruction Following Claude Sonnet 5 leads

Claude Sonnet 4.5: 75.0 (#78), Claude Sonnet 5: 76.3 (#41)

BenchmarkClaude Sonnet 4.5Claude Sonnet 5
LMArena Instruction Following14591452
IFEval85%—

Long Context Too close to call

Claude Sonnet 4.5: 45.2 (#46), Claude Sonnet 5: 44.8 (#55)

BenchmarkClaude Sonnet 4.5Claude Sonnet 5
LMArena Longer Query14761463

Writing & Preference Claude Sonnet 5 leads

Claude Sonnet 4.5: 66.5 (#34), Claude Sonnet 5: 69.2 (#25)

BenchmarkClaude Sonnet 4.5Claude Sonnet 5
LMArena Text14391442
LMArena Creative Writing14421416
EQ-Bench Creative Writing16781794
LMArena Multi-Turn14651454
WildBench85.4%—
EQ-Bench 4—1236

Frequently asked questions

Is Claude Sonnet 4.5 better than Claude Sonnet 5?

Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 44.1 on the Noometry Index.

Which is cheaper, Claude Sonnet 4.5 or Claude Sonnet 5?

Claude Sonnet 5 is cheaper. It lists at $2 per million input tokens and $10 per million output tokens; Claude Sonnet 4.5 lists at $3 and $15.

Is Claude Sonnet 4.5 or Claude Sonnet 5 better for coding?

Claude Sonnet 5 scores higher on coding benchmarks: 55.5 versus 47.3 in the Noometry coding category.

Which has the bigger context window?

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

How many benchmarks do Claude Sonnet 4.5 and Claude Sonnet 5 share?

41 benchmarks have published results for both models. Claude Sonnet 4.5 has 73 scored results on Noometry and Claude Sonnet 5 has 51.

Related comparisons

Go deeper