Log inSign up free
Benchmarks › Anthropic › Claude
Anthropicclosed weightsSupersededReasoning

Claude Opus 4.7

Adaptive thinking; scores below are the adaptive configuration.

Released
Apr 16, 2026
Price in / out
$5 / $25
Blended $/1M
$10
Context
1M

Benchmark scoresvs Claude Opus 4.6

BenchmarkScoreRank (current)Source
Terminal-Bench 2.1
68.5%#13 of 21benchlm.ai ↗
SWE-bench Verified
87.6%#2 of 9benchlm.ai ↗
GPQA Diamond
94.2%#3 of 11benchlm.ai ↗
Humanity's Last Exam
46.9%#6 of 15vellum.ai ↗
ARC-AGI-2
75.8%#7 of 10benchlm.ai ↗
τ²-bench (Telecom)
88.6%#6 of 10benchlm.ai ↗
Artificial Analysis Intelligence Index
53.5#2 of 11benchlm.ai ↗
LMArena Text
1490#1 of 5benchlm.ai ↗

Claude versions

Timeline
Version$ in / outTerminal-BenchSWE-benchGPQAHLEARC-AGI-2AIME
Claude 3.5 Sonnet (Oct 2024)Retired$3 / $15—49%————
Claude 3.7 SonnetRetired$3 / $15—62.3%————
Claude Opus 4Retired$15 / $75—72.5%————
Claude Sonnet 4Retired$3 / $15—72.7%————
Claude Opus 4.1Retired$15 / $75—74.5%————
Claude Sonnet 4.5Superseded$3 / $15—77.2%——13.6%—
Claude Haiku 4.5Superseded$1 / $543.8%73.3%—10.2%——
Claude Opus 4.5Superseded$5 / $25—80.9%—30.8%37.6%95.1%
Claude Opus 4.6Superseded$5 / $25—80.8%89.2%53%68.8%—
Claude Sonnet 4.6Superseded$3 / $1557.3%79.6%—49%58.3%—
Claude Opus 4.7Superseded$5 / $2568.5%87.6%94.2%46.9%75.8%—
Claude Opus 4.8Superseded$5 / $2571.9%88.6%93.6%57.9%72.1%—
Claude Fable 5Superseded$10 / $5080.5%95%—57.8%89.2%—
Claude Mythos 5Superseded$10 / $50—95.5%—64.5%——
Claude Sonnet 5Superseded$2 / $1074.5%85.2%—57.4%——
Claude Opus 5Superseded$5 / $2584.6%96%—56.6%90.4%—
Claude Fable 5.1$10 / $5085%——60.9%90%—
Claude Mythos 5.1$10 / $50——————
Claude Opus 5.5$4 / $2087.6%——67.7%91.7%—
Claude Sonnet 5.5$2 / $1083.1%——56.9%——
Claude Haiku 5.5$0.1 / $0.5———45.9%——