Noometry

Model comparison

Claude Opus 4.6 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 . 56 shared benchmarks.

Claude Opus 4.6 Anthropic

58.3

Rank #18 Confirmed

Claude Opus 4.8 Anthropic

60.7

Rank #9 Confirmed

Summary

  • They share 56 benchmarks with published results for both. Claude Opus 4.6 scores higher in 6 categories and Claude Opus 4.8 in 4 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in math, where Claude Opus 4.8 leads 78.2 to 62.7.
  • The biggest single-benchmark swing is FrontierMath Tier 4: 26.8% for Claude Opus 4.6 and 56.1% for Claude Opus 4.8.
  • Both cost about the same: $5 input and $25 output per million tokens.

Side by side

Claude Opus 4.6Claude Opus 4.8
ProviderAnthropicAnthropic
Noometry Index58.360.7
Released2026-02-042026-05-28
WeightsProprietaryProprietary
Context window1M1M
Max output128K128K
Input $ / M tokens$5$5
Output $ / M tokens$25$25
Results tracked6865

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

Category by category

Coding Claude Opus 4.8 leads

Claude Opus 4.6: 57.7 (#15), Claude Opus 4.8: 60.2 (#9)

BenchmarkClaude Opus 4.6Claude Opus 4.8
FrontierCode26.6%46.5%
LMArena WebDev15461556
GSO41.2%47.1%
WeirdML78%82.9%
LMArena Coding15351491
ALE-Bench996.51,564
SWE-bench Verified78.7%—
DeepSWE—59%
SWE-bench Verified (bash only)75.6%—
SWE-bench Multilingual72%—
SciCode—53.5%
AlgoTune1.47—

Agentic & Tool Use Claude Opus 4.6 leads

Claude Opus 4.6: 51.0 (#4), Claude Opus 4.8: 47.6 (#10)

BenchmarkClaude Opus 4.6Claude Opus 4.8
APEX-Agents46.3%48.9%
Remote Labor Index4.2%8.3%
τ²-bench Banking27.3%39.7%
DeepResearch Bench55.3%50.2%
GBAEval44.1%70.9%
LMArena Search12531204
Vending-Bench 28,0185,787
Terminal-Bench79.8%—
OSWorld 2.0—20.6%
Cybench93%—
PostTrainBench—33.8%
GDP.pdf—24%
METR Time Horizons78.9%—

Reasoning Claude Opus 4.8 leads

Claude Opus 4.6: 57.7 (#21), Claude Opus 4.8: 64.6 (#14)

BenchmarkClaude Opus 4.6Claude Opus 4.8
ARC-AGI-269.2%72.1%
SimpleBench67.6%64.8%
Kagi LLM Benchmark83.6%88.8%
NYT Connections (extended)92.1%91.1%
ARC-AGI-194%92.5%
Chess Puzzles17%34%
EnigmaEval7.6%23.5%
EBR-Bench12.7%28.6%
LMArena Hard Prompts15271481
Mystery Game Puzzles25%36%
DTBench91.2%94.9%
LMCA55.8%57.5%
Epoch Capabilities Index155.36158.3
ForecastBench6059.9
CritPt—20.9%
Thematic Generalization80.6%—
Surface Evolver Bench—87.5%
Bench to the Future 3—0.14

Math Claude Opus 4.8 leads

Claude Opus 4.6: 62.7 (#25), Claude Opus 4.8: 78.2 (#11)

BenchmarkClaude Opus 4.6Claude Opus 4.8
FrontierMath (Tiers 1-3)66%80%
FrontierMath Tier 426.8%56.1%
MathArena Final-Answer Competitions78.5%91.8%
OTIS Mock AIME 2024-202594.4%98.3%
ProofBench50%69%
LMArena Math15171487
FrontierMath (Feb 2025 set)40.7%47.2%
FrontierMath Tier 4 (v1)22.9%31.3%

Knowledge Too close to call

Claude Opus 4.6: 61.8 (#23), Claude Opus 4.8: 61.3 (#26)

BenchmarkClaude Opus 4.6Claude Opus 4.8
GPQA Diamond90.5%91%
SimpleQA Verified47%53%
LMArena Expert15461501
Humanity's Last Exam34.4%—
Vectara Hallucination Rate12.2%—

Multimodal Claude Opus 4.8 leads

Claude Opus 4.6: 37.0 (#67), Claude Opus 4.8: 43.2 (#20)

BenchmarkClaude Opus 4.6Claude Opus 4.8
LMArena Vision13161295
Furniture Assembly28.3%42.5%
LMArena Document15071475
Blueprint-Bench 2—14.5%

Multilingual Claude Opus 4.6 leads

Claude Opus 4.6: 58.2 (#4), Claude Opus 4.8: 55.4 (#31)

BenchmarkClaude Opus 4.6Claude Opus 4.8
LMArena Non-English14901450
LMArena Chinese15511509
LMArena French15131482
LMArena German15041470
LMArena Japanese14841440
LMArena Korean14651432
LMArena Russian14971476
LMArena Spanish15121470

Instruction Following Claude Opus 4.6 leads

Claude Opus 4.6: 79.7 (#3), Claude Opus 4.8: 77.6 (#22)

BenchmarkClaude Opus 4.6Claude Opus 4.8
LMArena Instruction Following15231476

Long Context Claude Opus 4.6 leads

Claude Opus 4.6: 48.1 (#13), Claude Opus 4.8: 45.4 (#37)

BenchmarkClaude Opus 4.6Claude Opus 4.8
LMArena Longer Query15201482
CL-bench20.7%—
CL-bench Life17%—

Writing & Preference Claude Opus 4.6 leads

Claude Opus 4.6: 73.7 (#9), Claude Opus 4.8: 72.1 (#17)

BenchmarkClaude Opus 4.6Claude Opus 4.8
LMArena Text15041460
LMArena Creative Writing15061453
EQ-Bench Creative Writing18091840
EQ-Bench 412231281
LMArena Multi-Turn15141476

Frequently asked questions

Is Claude Opus 4.6 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 4.6 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 4.6 lists at $5 and $25.

Is Claude Opus 4.6 or Claude Opus 4.8 better for coding?

Claude Opus 4.8 scores higher on coding benchmarks: 60.2 versus 57.7 in the Noometry coding category.

Which has the bigger context window?

Both accept 1M tokens.

How many benchmarks do Claude Opus 4.6 and Claude Opus 4.8 share?

56 benchmarks have published results for both models. Claude Opus 4.6 has 68 scored results on Noometry and Claude Opus 4.8 has 65.

Related comparisons

Go deeper