Noometry

Model comparison

Claude Opus 4.7 vs Claude Opus 4.8

Claude Opus 4.8 is the stronger model overall, scoring 60.7 to 58.3 on the Noometry Index.

Last verified . 59 shared benchmarks.

Claude Opus 4.7 Anthropic

58.3

Rank #19 Confirmed

Claude Opus 4.8 Anthropic

60.7

Rank #13 Confirmed

Summary

  • They share 59 benchmarks with published results for both. Claude Opus 4.7 scores higher in 6 categories and Claude Opus 4.8 in 4 categories; 6 gaps are clear of the uncertainty.
  • The widest gap is in math, where Claude Opus 4.8 leads 78.4 to 66.7.
  • The biggest single-benchmark swing is NYT Connections (extended): 39% for Claude Opus 4.7 and 91.1% for Claude Opus 4.8.
  • Both cost about the same: $5 input and $25 output per million tokens.

Side by side

Claude Opus 4.7Claude Opus 4.8
ProviderAnthropicAnthropic
Noometry Index58.360.7
Released2026-04-142026-05-28
WeightsProprietaryProprietary
Context window1M1M
Max output128K128K
Input $ / M tokens$5$5
Output $ / M tokens$25$25
Results tracked6665

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

Category by category

Coding Too close to call

Claude Opus 4.7: 59.6 (#13), Claude Opus 4.8: 59.9 (#12)

BenchmarkClaude Opus 4.7Claude Opus 4.8
FrontierCode38.5%46.5%
LMArena WebDev15581556
SciCode54.5%53.5%
GSO44.1%47.1%
WeirdML76.4%82.9%
LMArena Coding15181490
ALE-Bench1,3231,564
SWE-bench Verified83.5%—
DeepSWE—59%
MirrorCode31.1%—

Agentic & Tool Use Too close to call

Claude Opus 4.7: 47.9 (#10), Claude Opus 4.8: 47.6 (#11)

BenchmarkClaude Opus 4.7Claude Opus 4.8
APEX-Agents49.2%48.9%
OSWorld 2.018.2%20.6%
τ²-bench Banking40.2%39.7%
PostTrainBench28.6%33.8%
GBAEval43.8%70.9%
GDP.pdf21%24%
LMArena Search12331204
Vending-Bench 210,9375,787
Terminal-Bench80.2%—
Remote Labor Index—8.3%
DeepResearch Bench—50.2%
ExploitBench26.5%—

Reasoning Claude Opus 4.8 leads

Claude Opus 4.7: 53.8 (#29), Claude Opus 4.8: 64.7 (#16)

BenchmarkClaude Opus 4.7Claude Opus 4.8
ARC-AGI-275.8%72.1%
SimpleBench61.7%64.8%
Kagi LLM Benchmark80.7%88.8%
NYT Connections (extended)39%91.1%
ARC-AGI-193.5%92.5%
CritPt12%20.9%
Chess Puzzles30%34%
EBR-Bench19%28.6%
LMArena Hard Prompts15061482
Mystery Game Puzzles28%36%
DTBench94.7%94.9%
LMCA52.2%57.5%
Epoch Capabilities Index156.25158.21
ForecastBench60.359.9
EnigmaEval—23.5%
Thematic Generalization72.8%—
Surface Evolver Bench—87.5%
Bench to the Future 3—0.14

Math Claude Opus 4.8 leads

Claude Opus 4.7: 66.7 (#26), Claude Opus 4.8: 78.4 (#13)

BenchmarkClaude Opus 4.7Claude Opus 4.8
FrontierMath (Tiers 1-3)70.2%80%
FrontierMath Tier 431.7%56.1%
MathArena Final-Answer Competitions73.6%91.8%
OTIS Mock AIME 2024-202597.8%98.3%
ProofBench54%69%
LMArena Math14991487
FrontierMath (Feb 2025 set)43.8%47.2%
FrontierMath Tier 4 (v1)22.9%31.3%

Knowledge Claude Opus 4.7 leads

Claude Opus 4.7: 62.6 (#23), Claude Opus 4.8: 61.3 (#29)

BenchmarkClaude Opus 4.7Claude Opus 4.8
GPQA Diamond90.2%91%
SimpleQA Verified51.7%53%
LMArena Expert15211502
Humanity's Last Exam36.2%—
Vectara Hallucination Rate12%—

Multimodal Claude Opus 4.8 leads

Claude Opus 4.7: 41.2 (#38), Claude Opus 4.8: 42.9 (#26)

BenchmarkClaude Opus 4.7Claude Opus 4.8
LMArena Vision13161294
Blueprint-Bench 224.5%14.5%
Furniture Assembly33.3%42.5%
LMArena Document14951475

Multilingual Claude Opus 4.7 leads

Claude Opus 4.7: 57.3 (#10), Claude Opus 4.8: 55.2 (#33)

BenchmarkClaude Opus 4.7Claude Opus 4.8
LMArena Non-English14801450
LMArena Chinese15311507
LMArena French15031481
LMArena German14951472
LMArena Japanese14721440
LMArena Korean14641432
LMArena Russian14941474
LMArena Spanish14951466

Instruction Following Too close to call

Claude Opus 4.7: 78.4 (#10), Claude Opus 4.8: 77.4 (#24)

BenchmarkClaude Opus 4.7Claude Opus 4.8
LMArena Instruction Following14981476

Long Context Too close to call

Claude Opus 4.7: 46.2 (#25), Claude Opus 4.8: 45.4 (#35)

BenchmarkClaude Opus 4.7Claude Opus 4.8
LMArena Longer Query15051483

Writing & Preference Claude Opus 4.7 leads

Claude Opus 4.7: 75.1 (#8), Claude Opus 4.8: 72.0 (#16)

BenchmarkClaude Opus 4.7Claude Opus 4.8
LMArena Text14901461
LMArena Creative Writing14861454
EQ-Bench Creative Writing19141840
EQ-Bench 413111281
LMArena Multi-Turn15051476

Frequently asked questions

Is Claude Opus 4.7 better than Claude Opus 4.8?

Claude Opus 4.8 is the stronger model overall, scoring 60.7 to 58.3 on the Noometry Index.

Which is cheaper, Claude Opus 4.7 or Claude Opus 4.8?

Claude Opus 4.8 is cheaper. It lists at $5 per million input tokens and $25 per million output tokens; Claude Opus 4.7 lists at $5 and $25.

Is Claude Opus 4.7 or Claude Opus 4.8 better for coding?

They score almost the same on coding (59.6 vs 59.9); test both on your own repository before choosing.

Which has the bigger context window?

Both accept 1M tokens.

How many benchmarks do Claude Opus 4.7 and Claude Opus 4.8 share?

59 benchmarks have published results for both models. Claude Opus 4.7 has 66 scored results on Noometry and Claude Opus 4.8 has 65.

Related comparisons

Go deeper