Noometry

Model comparison

Claude Sonnet 4.5 vs GPT-5 Nano

Claude Sonnet 4.5 is the stronger model overall, scoring 44.0 to 33.4 on the Noometry Index. GPT-5 Nano costs 44× less per token, which makes it the better buy when Claude Sonnet 4.5's lead doesn't matter for your workload.

Last verified . 46 shared benchmarks.

Claude Sonnet 4.5 Anthropic

44.0

Rank #76 Confirmed

GPT-5 Nano OpenAI

33.4

Rank #234 Confirmed

Summary

  • They share 46 benchmarks with published results for both. Claude Sonnet 4.5 scores higher in 9 categories and GPT-5 Nano in 1 category; 9 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where Claude Sonnet 4.5 leads 66.7 to 39.4.
  • The biggest single-benchmark swing is ARC-AGI-1: 63.7% for Claude Sonnet 4.5 and 20.7% for GPT-5 Nano.
  • GPT-5 Nano is cheaper at $0.05 / $0.40 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.5.
  • Claude Sonnet 4.5 accepts more context: 1M tokens versus 400K.

Side by side

Claude Sonnet 4.5GPT-5 Nano
ProviderAnthropicOpenAI
Noometry Index44.033.4
Released2025-09-292025-08-07
WeightsProprietaryProprietary
Context window1M400K
Max output64K128K
Input $ / M tokens$3$0.05
Output $ / M tokens$15$0.40
Results tracked7348

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

Category by category

Coding Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 47.8 (#54), GPT-5 Nano: 33.3 (#248)

BenchmarkClaude Sonnet 4.5GPT-5 Nano
SWE-bench Verified (bash only)71.4%34.8%
WeirdML47.7%38.1%
LMArena Coding14891351
ALE-Bench796.15718.67
SWE-bench Verified71.3%—
LMArena WebDev1393—
SWE-bench Multilingual67%—
SciCode44.7%—
GSO14.7%—
AlgoTune1.52—

Agentic & Tool Use Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 38.1 (#30), GPT-5 Nano: 25.6 (#99)

BenchmarkClaude Sonnet 4.5GPT-5 Nano
Terminal-Bench46.5%21.8%
Berkeley Function Calling Leaderboard73.2%51.5%
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%—
LMArena Search1159—
METR Time Horizons67.4%—
Vending-Bench 23,839—

Reasoning Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 26.9 (#113), GPT-5 Nano: 16.4 (#291)

BenchmarkClaude Sonnet 4.5GPT-5 Nano
ARC-AGI-213.6%2.6%
Kagi LLM Benchmark57.9%62.2%
ARC-AGI-163.7%20.7%
Chess Puzzles12%27%
LMArena Hard Prompts14621329
Mystery Game Puzzles17%9%
DTBench83.2%62.7%
LMCA38.8%7.9%
Epoch Capabilities Index146.84139.39
ForecastBench61.959.1
SimpleBench54.3%—
NYT Connections (extended)37.3%—
CritPt1.1%—
EnigmaEval6%—
EBR-Bench2.4%—

Math Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 31.9 (#190), GPT-5 Nano: 29.1 (#221)

BenchmarkClaude Sonnet 4.5GPT-5 Nano
FrontierMath (Tiers 1-3)23.9%20%
FrontierMath Tier 42.4%2.4%
OTIS Mock AIME 2024-202577.8%81.1%
ProofBench19%12%
Omni-MATH55.3%54.6%
LMArena Math14481318
MATH Level 597.7%95.2%
FrontierMath (Feb 2025 set)15.2%8.3%
FrontierMath Tier 4 (v1)4.2%2.1%

Knowledge Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 48.4 (#70), GPT-5 Nano: 35.8 (#166)

BenchmarkClaude Sonnet 4.5GPT-5 Nano
GPQA Diamond82.3%69.4%
SimpleQA Verified30.7%11.7%
MMLU-Pro86.9%77.8%
Vectara Hallucination Rate12%10.5%
GPQA (HELM)68.6%67.9%
LMArena Expert14831321
Humanity's Last Exam13.7%—

Multimodal Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 34.1 (#84), GPT-5 Nano: 30.7 (#102)

BenchmarkClaude Sonnet 4.5GPT-5 Nano
VPCT39.8%37.2%
LMArena Vision—1159
LMArena Document1450—

Multilingual Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 53.6 (#65), GPT-5 Nano: 45.5 (#167)

BenchmarkClaude Sonnet 4.5GPT-5 Nano
LMArena Non-English14251313
LMArena Chinese14591355
LMArena German14271329
LMArena Korean14031269
LMArena Russian14361296
LMArena Spanish14571360
LMArena French1459—
LMArena Japanese1390—

Instruction Following Too close to call

Claude Sonnet 4.5: 75.2 (#74), GPT-5 Nano: 75.2 (#72)

BenchmarkClaude Sonnet 4.5GPT-5 Nano
IFEval85%93.2%
LMArena Instruction Following14591306

Long Context Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 45.2 (#45), GPT-5 Nano: 31.3 (#275)

BenchmarkClaude Sonnet 4.5GPT-5 Nano
LMArena Longer Query14761313
Fiction.LiveBench—44.4%

Writing & Preference Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 66.7 (#33), GPT-5 Nano: 39.4 (#244)

BenchmarkClaude Sonnet 4.5GPT-5 Nano
LMArena Text14391320
LMArena Creative Writing14421250
EQ-Bench Creative Writing1678705
WildBench85.4%80.6%
LMArena Multi-Turn14651311

Frequently asked questions

Is Claude Sonnet 4.5 better than GPT-5 Nano?

Claude Sonnet 4.5 is the stronger model overall, scoring 44.0 to 33.4 on the Noometry Index. GPT-5 Nano costs 44× less per token, which makes it the better buy when Claude Sonnet 4.5's lead doesn't matter for your workload.

Which is cheaper, Claude Sonnet 4.5 or GPT-5 Nano?

GPT-5 Nano is cheaper. It lists at $0.05 per million input tokens and $0.40 per million output tokens; Claude Sonnet 4.5 lists at $3 and $15.

Is Claude Sonnet 4.5 or GPT-5 Nano better for coding?

Claude Sonnet 4.5 scores higher on coding benchmarks: 47.8 versus 33.3 in the Noometry coding category.

Which has the bigger context window?

Claude Sonnet 4.5 does, with 1M tokens against 400K.

How many benchmarks do Claude Sonnet 4.5 and GPT-5 Nano share?

46 benchmarks have published results for both models. Claude Sonnet 4.5 has 73 scored results on Noometry and GPT-5 Nano has 48.

Related comparisons

Go deeper