Model comparison

Claude Sonnet 4.5 vs GPT-4o

Claude Sonnet 4.5 is the stronger model overall, scoring 44.1 to 28.6 on the Noometry Index.

Last verified October 10, 2026. 49 shared benchmarks.

Claude Sonnet 4.5 Anthropic

44.1

Rank #81 Confirmed

GPT-4o OpenAI

28.6

Rank #324 Confirmed

Summary

They share 49 benchmarks with published results for both. Claude Sonnet 4.5 scores higher in 10 categories and GPT-4o in 0 categories; 9 gaps are clear of the uncertainty.
The widest gap is in coding, where Claude Sonnet 4.5 leads 47.3 to 24.8.
The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 77.8% for Claude Sonnet 4.5 and 6.4% for GPT-4o.
GPT-4o is cheaper at $2.50 / $10 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.5.
Claude Sonnet 4.5 accepts more context: 200K tokens versus 128K.

Side by side

Claude Sonnet 4.5 and GPT-4o specifications
	Claude Sonnet 4.5	GPT-4o
Provider	Anthropic	OpenAI
Noometry Index	44.1	28.6
Released	2025-09-29	2024-05-13
Weights	Proprietary	Proprietary
Context window	200K	128K
Max output	64K	16K
Input $ / M tokens	$3	$2.50
Output $ / M tokens	$15	$10
Results tracked	73	72

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-4o: 24.8 (#328)

Coding benchmarks
Benchmark	Claude Sonnet 4.5	GPT-4o
SWE-bench Verified	71.3%	31%
SWE-bench Verified (bash only)	71.4%	21.6%
GSO	14.7%	0%
WeirdML	47.7%	25.1%
LMArena Coding	1489	1297
Aider Polyglot	—	45.3%
LMArena WebDev	1393	—
SWE-bench Multilingual	67%	—
SciCode	44.7%	—
BigCodeBench Instruct	—	51.1%
LiveBench Coding	—	51.4%
BigCodeBench Complete	—	61.1%
CadEval	—	26%
ALE-Bench	796.15	—
AlgoTune	1.52	—
HumanEval+	—	87.2%
MBPP+	—	72.2%

Agentic & Tool Use Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 38.3 (#32), GPT-4o: 21.0 (#141)

Agentic & Tool Use benchmarks
Benchmark	Claude Sonnet 4.5	GPT-4o
GDPval	42.5%	9.9%
Cybench	60%	12.5%
LMArena Search	1159	1006
METR Time Horizons	67.4%	40.8%
Terminal-Bench	46.5%	—
Berkeley Function Calling Leaderboard	73.2%	—
Remote Labor Index	2.1%	—
TheAgentCompany	—	8.6%
τ²-bench Airline	72%	—
τ²-bench Banking	25.3%	—
τ²-bench Retail	72.4%	—
τ²-bench Telecom	84.9%	—
DeepResearch Bench	52.6%	—
OSWorld	62.9%	—
BALROG	—	32.3%
Vending-Bench 2	3,839	—

Reasoning Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 26.9 (#125), GPT-4o: 9.4 (#343)

Reasoning benchmarks
Benchmark	Claude Sonnet 4.5	GPT-4o
ARC-AGI-2	13.6%	0%
SimpleBench	54.3%	17.8%
ARC-AGI-1	63.7%	4.5%
CritPt	1.1%	0%
Chess Puzzles	12%	13%
EnigmaEval	6%	0.8%
LMArena Hard Prompts	1462	1281
DTBench	83.2%	64.5%
LMCA	38.8%	16.6%
Epoch Capabilities Index	146.84	128.97
ForecastBench	61.9	57.7
Kagi LLM Benchmark	57.9%	—
NYT Connections (extended)	37.3%	—
EBR-Bench	2.4%	—
LiveBench Reasoning	—	55.8%
Mystery Game Puzzles	17%	—
LiveBench Data Analysis	—	60.9%
LiveBench	—	55.3%

Math Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 32.3 (#216), GPT-4o: 10.6 (#312)

Math benchmarks
Benchmark	Claude Sonnet 4.5	GPT-4o
FrontierMath (Tiers 1-3)	23.9%	0.4%
OTIS Mock AIME 2024-2025	77.8%	6.4%
Omni-MATH	55.3%	29.3%
LMArena Math	1449	1285
MATH Level 5	97.7%	53.3%
FrontierMath (Feb 2025 set)	15.2%	0.3%
FrontierMath Tier 4	2.4%	—
ProofBench	19%	—
LiveBench Math	—	49.5%
FrontierMath Tier 4 (v1)	4.2%	—

Knowledge Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 48.4 (#76), GPT-4o: 28.8 (#242)

Knowledge benchmarks
Benchmark	Claude Sonnet 4.5	GPT-4o
GPQA Diamond	82.3%	49.2%
Humanity's Last Exam	13.7%	2.7%
SimpleQA Verified	30.7%	26%
MMLU-Pro	86.9%	71.3%
Vectara Hallucination Rate	12%	9.6%
GPQA (HELM)	68.6%	52%
LMArena Expert	1482	1250
Confabulations	—	15.3%
MMLU	—	88.1%

Multimodal Too close to call

Claude Sonnet 4.5: 34.8 (#89), GPT-4o: 34.5 (#91)

Multimodal benchmarks
Benchmark	Claude Sonnet 4.5	GPT-4o
VPCT	39.8%	40%
LMArena Vision	—	1137
Video-MME	—	71.9%
GeoBench	—	71%
LMArena Document	1450	—
ScienceQA	—	88.5%

Multilingual Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 53.4 (#69), GPT-4o: 43.2 (#186)

Multilingual benchmarks
Benchmark	Claude Sonnet 4.5	GPT-4o
LMArena Non-English	1425	1283
LMArena Chinese	1459	1277
LMArena French	1458	1304
LMArena German	1427	1282
LMArena Japanese	1390	1257
LMArena Korean	1403	1234
LMArena Russian	1437	1286
LMArena Spanish	1457	1292

Instruction Following Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 75.0 (#78), GPT-4o: 66.6 (#207)

Instruction Following benchmarks
Benchmark	Claude Sonnet 4.5	GPT-4o
IFEval	85%	81.7%
LMArena Instruction Following	1459	1278
LiveBench Instruction Following	—	68.6%

Long Context Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 45.2 (#46), GPT-4o: 39.4 (#179)

Long Context benchmarks
Benchmark	Claude Sonnet 4.5	GPT-4o
LMArena Longer Query	1476	1289
Fiction.LiveBench	—	66.7%

Writing & Preference Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 66.5 (#34), GPT-4o: 52.6 (#166)

Writing & Preference benchmarks
Benchmark	Claude Sonnet 4.5	GPT-4o
LMArena Text	1439	1300
LMArena Creative Writing	1442	1292
WildBench	85.4%	82.8%
LMArena Multi-Turn	1465	1302
Short-Story Creative Writing	—	81.8%
EQ-Bench Creative Writing	1678	—
LiveBench Language	—	47.6%

Frequently asked questions

Is Claude Sonnet 4.5 better than GPT-4o?

Claude Sonnet 4.5 is the stronger model overall, scoring 44.1 to 28.6 on the Noometry Index.

Which is cheaper, Claude Sonnet 4.5 or GPT-4o?

GPT-4o is cheaper. It lists at $2.50 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 GPT-4o better for coding?

Claude Sonnet 4.5 scores higher on coding benchmarks: 47.3 versus 24.8 in the Noometry coding category.

Which has the bigger context window?

Claude Sonnet 4.5 does, with 200K tokens against 128K.

How many benchmarks do Claude Sonnet 4.5 and GPT-4o share?

49 benchmarks have published results for both models. Claude Sonnet 4.5 has 73 scored results on Noometry and GPT-4o has 72.