Noometry

Model comparison

Claude Sonnet 4.5 vs GPT-5 Mini

Claude Sonnet 4.5 is the stronger model overall, scoring 44.1 to 41.8 on the Noometry Index. GPT-5 Mini costs 8.7× less per token, which makes it the better buy when Claude Sonnet 4.5's lead doesn't matter for your workload.

Last verified . 56 shared benchmarks.

Claude Sonnet 4.5 Anthropic

44.1

Rank #81 Confirmed

GPT-5 Mini OpenAI

41.8

Rank #128 Confirmed

Summary

  • They share 56 benchmarks with published results for both. Claude Sonnet 4.5 scores higher in 7 categories and GPT-5 Mini in 3 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in math, where GPT-5 Mini leads 46.7 to 32.3.
  • The biggest single-benchmark swing is SWE-bench Multilingual: 67% for Claude Sonnet 4.5 and 39.7% for GPT-5 Mini.
  • GPT-5 Mini is cheaper at $0.25 / $2 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.5.
  • GPT-5 Mini accepts more context: 400K tokens versus 200K.

Side by side

Claude Sonnet 4.5GPT-5 Mini
ProviderAnthropicOpenAI
Noometry Index44.141.8
Released2025-09-292025-08-07
WeightsProprietaryProprietary
Context window200K400K
Max output64K128K
Input $ / M tokens$3$0.25
Output $ / M tokens$15$2
Results tracked7360

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: 47.3 (#61), GPT-5 Mini: 40.1 (#146)

BenchmarkClaude Sonnet 4.5GPT-5 Mini
SWE-bench Verified71.3%64.7%
SWE-bench Verified (bash only)71.4%59.8%
SWE-bench Multilingual67%39.7%
SciCode44.7%39.2%
WeirdML47.7%52.7%
LMArena Coding14891406
ALE-Bench796.15799.77
AlgoTune1.521.38
LMArena WebDev1393—
GSO14.7%—

Agentic & Tool Use Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 38.3 (#32), GPT-5 Mini: 31.1 (#70)

BenchmarkClaude Sonnet 4.5GPT-5 Mini
Terminal-Bench46.5%34.8%
Berkeley Function Calling Leaderboard73.2%55.5%
Vending-Bench 23,839-31.18
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%—

Reasoning Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 26.9 (#125), GPT-5 Mini: 23.9 (#168)

BenchmarkClaude Sonnet 4.5GPT-5 Mini
ARC-AGI-213.6%4.4%
Kagi LLM Benchmark57.9%70.3%
ARC-AGI-163.7%54.3%
CritPt1.1%0%
Chess Puzzles12%30%
EnigmaEval6%8.2%
LMArena Hard Prompts14621380
Mystery Game Puzzles17%10%
DTBench83.2%80.5%
LMCA38.8%34.2%
Epoch Capabilities Index146.84145.52
ForecastBench61.961
SimpleBench54.3%—
NYT Connections (extended)37.3%—
EBR-Bench2.4%—

Math GPT-5 Mini leads

Claude Sonnet 4.5: 32.3 (#216), GPT-5 Mini: 46.7 (#69)

BenchmarkClaude Sonnet 4.5GPT-5 Mini
FrontierMath (Tiers 1-3)23.9%46.7%
FrontierMath Tier 42.4%12.2%
OTIS Mock AIME 2024-202577.8%86.7%
ProofBench19%9%
Omni-MATH55.3%72.2%
LMArena Math14491378
MATH Level 597.7%97.8%
FrontierMath (Feb 2025 set)15.2%27.2%
FrontierMath Tier 4 (v1)4.2%6.3%

Knowledge Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 48.4 (#76), GPT-5 Mini: 45.6 (#86)

BenchmarkClaude Sonnet 4.5GPT-5 Mini
GPQA Diamond82.3%75%
Humanity's Last Exam13.7%19.4%
SimpleQA Verified30.7%21.6%
MMLU-Pro86.9%83.5%
Vectara Hallucination Rate12%12.9%
GPQA (HELM)68.6%75.6%
LMArena Expert14821379
Confabulations—13.3%

Multimodal Too close to call

Claude Sonnet 4.5: 34.8 (#89), GPT-5 Mini: 35.6 (#85)

BenchmarkClaude Sonnet 4.5GPT-5 Mini
VPCT39.8%40.2%
LMArena Vision—1202
LMArena Document1450—

Multilingual Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 53.4 (#69), GPT-5 Mini: 48.9 (#137)

BenchmarkClaude Sonnet 4.5GPT-5 Mini
LMArena Non-English14251363
LMArena Chinese14591385
LMArena French14581386
LMArena German14271366
LMArena Japanese13901341
LMArena Korean14031308
LMArena Russian14371362
LMArena Spanish14571355

Instruction Following GPT-5 Mini leads

Claude Sonnet 4.5: 75.0 (#78), GPT-5 Mini: 76.2 (#46)

BenchmarkClaude Sonnet 4.5GPT-5 Mini
IFEval85%92.7%
LMArena Instruction Following14591357

Long Context Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 45.2 (#46), GPT-5 Mini: 41.9 (#132)

BenchmarkClaude Sonnet 4.5GPT-5 Mini
LMArena Longer Query14761355
Fiction.LiveBench—69.4%

Writing & Preference Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 66.5 (#34), GPT-5 Mini: 55.2 (#148)

BenchmarkClaude Sonnet 4.5GPT-5 Mini
LMArena Text14391373
LMArena Creative Writing14421325
EQ-Bench Creative Writing16781313
WildBench85.4%85.5%
LMArena Multi-Turn14651363
Short-Story Creative Writing—83.1%

Frequently asked questions

Is Claude Sonnet 4.5 better than GPT-5 Mini?

Claude Sonnet 4.5 is the stronger model overall, scoring 44.1 to 41.8 on the Noometry Index. GPT-5 Mini costs 8.7× 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 Mini?

GPT-5 Mini is cheaper. It lists at $0.25 per million input tokens and $2 per million output tokens; Claude Sonnet 4.5 lists at $3 and $15.

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

Claude Sonnet 4.5 scores higher on coding benchmarks: 47.3 versus 40.1 in the Noometry coding category.

Which has the bigger context window?

GPT-5 Mini does, with 400K tokens against 200K.

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

56 benchmarks have published results for both models. Claude Sonnet 4.5 has 73 scored results on Noometry and GPT-5 Mini has 60.

Related comparisons

Go deeper