Noometry

Model comparison

Claude Opus 4.7 vs GPT-5

Claude Opus 4.7 is the stronger model overall, scoring 58.3 to 50.9 on the Noometry Index. GPT-5 costs 2.9× less per token, which makes it the better buy when Claude Opus 4.7's lead doesn't matter for your workload.

Last verified . 49 shared benchmarks.

Claude Opus 4.7 Anthropic

58.3

Rank #19 Confirmed

GPT-5 OpenAI

50.9

Rank #45 Confirmed

Summary

  • They share 49 benchmarks with published results for both. Claude Opus 4.7 scores higher in 8 categories and GPT-5 in 2 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in long context, where GPT-5 leads 69.5 to 46.2.
  • The biggest single-benchmark swing is ARC-AGI-2: 75.8% for Claude Opus 4.7 and 9.9% for GPT-5.
  • GPT-5 is cheaper at $1.25 / $10 per million input/output tokens, against $5 / $25 for Claude Opus 4.7.
  • Claude Opus 4.7 accepts more context: 1M tokens versus 400K.

Side by side

Claude Opus 4.7GPT-5
ProviderAnthropicOpenAI
Noometry Index58.350.9
Released2026-04-142025-08-07
WeightsProprietaryProprietary
Context window1M400K
Max output128K128K
Input $ / M tokens$5$1.25
Output $ / M tokens$25$10
Results tracked6669

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), GPT-5: 50.3 (#47)

BenchmarkClaude Opus 4.7GPT-5
SWE-bench Verified83.5%73.6%
LMArena WebDev15581418
SciCode54.5%42.9%
GSO44.1%6.9%
WeirdML76.4%60.7%
LMArena Coding15181436
ALE-Bench1,3231,162
FrontierCode38.5%—
SWE-bench Verified (bash only)—65%
Aider Polyglot—88%
MirrorCode31.1%—
AlgoTune—1.67

Agentic & Tool Use Claude Opus 4.7 leads

Claude Opus 4.7: 47.9 (#10), GPT-5: 33.1 (#56)

BenchmarkClaude Opus 4.7GPT-5
Terminal-Bench80.2%49.6%
LMArena Search12331133
APEX-Agents49.2%—
OSWorld 2.018.2%—
GDPval—34.8%
Remote Labor Index—1.7%
τ²-bench Banking40.2%—
DeepResearch Bench—49.6%
PostTrainBench28.6%—
BALROG—32.8%
ExploitBench26.5%—
GBAEval43.8%—
GDP.pdf21%—
METR Time Horizons—69.6%
Vending-Bench 210,937—

Reasoning Claude Opus 4.7 leads

Claude Opus 4.7: 53.8 (#29), GPT-5: 38.3 (#64)

BenchmarkClaude Opus 4.7GPT-5
ARC-AGI-275.8%9.9%
SimpleBench61.7%56.7%
Kagi LLM Benchmark80.7%72.7%
ARC-AGI-193.5%65.7%
CritPt12%12.6%
Chess Puzzles30%37%
EBR-Bench19%12.7%
LMArena Hard Prompts15061416
Mystery Game Puzzles28%23%
DTBench94.7%90.7%
LMCA52.2%40%
Epoch Capabilities Index156.25150
ForecastBench60.361.4
NYT Connections (extended)39%—
EnigmaEval—10.5%
Thematic Generalization72.8%—

Math Claude Opus 4.7 leads

Claude Opus 4.7: 66.7 (#26), GPT-5: 55.0 (#44)

BenchmarkClaude Opus 4.7GPT-5
FrontierMath (Tiers 1-3)70.2%55.4%
FrontierMath Tier 431.7%22%
OTIS Mock AIME 2024-202597.8%91.4%
ProofBench54%18%
LMArena Math14991407
FrontierMath (Feb 2025 set)43.8%32.4%
FrontierMath Tier 4 (v1)22.9%12.5%
MathArena Final-Answer Competitions73.6%—
Omni-MATH—64.7%
MATH Level 5—98.1%

Knowledge Claude Opus 4.7 leads

Claude Opus 4.7: 62.6 (#23), GPT-5: 56.6 (#43)

BenchmarkClaude Opus 4.7GPT-5
GPQA Diamond90.2%86.2%
Humanity's Last Exam36.2%25.3%
SimpleQA Verified51.7%50.1%
Vectara Hallucination Rate12%14.7%
LMArena Expert15211419
MMLU-Pro—86.3%
Confabulations—10.3%
GPQA (HELM)—79.2%

Multimodal GPT-5 leads

Claude Opus 4.7: 41.2 (#38), GPT-5: 46.8 (#13)

BenchmarkClaude Opus 4.7GPT-5
LMArena Vision13161232
GeoBench—81%
VPCT—66%
Blueprint-Bench 224.5%—
Furniture Assembly33.3%—
LMArena Document1495—

Multilingual Claude Opus 4.7 leads

Claude Opus 4.7: 57.3 (#10), GPT-5: 51.4 (#110)

BenchmarkClaude Opus 4.7GPT-5
LMArena Non-English14801397
LMArena Chinese15311422
LMArena French15031410
LMArena German14951416
LMArena Japanese14721409
LMArena Korean14641360
LMArena Russian14941406
LMArena Spanish14951399

Instruction Following Claude Opus 4.7 leads

Claude Opus 4.7: 78.4 (#10), GPT-5: 73.8 (#113)

BenchmarkClaude Opus 4.7GPT-5
LMArena Instruction Following14981388
IFEval—87.5%

Long Context GPT-5 leads

Claude Opus 4.7: 46.2 (#25), GPT-5: 69.5 (#2)

BenchmarkClaude Opus 4.7GPT-5
LMArena Longer Query15051399
Fiction.LiveBench—97.2%

Writing & Preference Claude Opus 4.7 leads

Claude Opus 4.7: 75.1 (#8), GPT-5: 63.4 (#65)

BenchmarkClaude Opus 4.7GPT-5
LMArena Text14901406
LMArena Creative Writing14861365
EQ-Bench Creative Writing19141627
LMArena Multi-Turn15051426
Short-Story Creative Writing—86%
WildBench—85.7%
EQ-Bench 41311—

Frequently asked questions

Is Claude Opus 4.7 better than GPT-5?

Claude Opus 4.7 is the stronger model overall, scoring 58.3 to 50.9 on the Noometry Index. GPT-5 costs 2.9× 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 GPT-5?

GPT-5 is cheaper. It lists at $1.25 per million input tokens and $10 per million output tokens; Claude Opus 4.7 lists at $5 and $25.

Is Claude Opus 4.7 or GPT-5 better for coding?

Claude Opus 4.7 scores higher on coding benchmarks: 59.6 versus 50.3 in the Noometry coding category.

Which has the bigger context window?

Claude Opus 4.7 does, with 1M tokens against 400K.

How many benchmarks do Claude Opus 4.7 and GPT-5 share?

49 benchmarks have published results for both models. Claude Opus 4.7 has 66 scored results on Noometry and GPT-5 has 69.

Related comparisons

Go deeper