Claude Opus 4.5 vs Claude Sonnet 4.6

Verdict

Claude Sonnet 4.6 leads 9–5 across 15 shared benchmarks.

Claude Opus 4.5 5 · Claude Sonnet 4.6 9 · 1 tied · higher isn't always better — check the caveats below.

Per-benchmark head-to-head

Reported scores — protocols may differ. How we compare

Sort

Claude Opus 4.5Claude Sonnet 4.6

80
BARC-AGI
86.5
37.64
AARC-AGI-2
60.42
9.51
AChess Puzzles
18.98
54.8
BDeepResearch Bench
54.9
83.12
ADTBench
83.12
81.4
BGPQA diamond
83.16
52.31
ALMCA
54.72
14.08
AMystery Game Puzzles
7.47
66.3
BOSWorld
72.1
86.1
BOTIS Mock AIME 2024-2025
85.78
36
BProofBench
45
45.7
ASimpleQA Verified
35.5
76.65
BSWE-Bench verified
75.21
63.15
BTerminal Bench
53.43
63.74
AWeirdML
66.07

Capabilities

Reasoning

Reasoning: Claude Sonnet 4.6 leads 3–1 across 4 shared reasoning benchmarks (largest gap: SimpleQA Verified, 45.7 vs 35.5 — protocols undisclosed).

2 reproduced6 unverified

Unknown

Unknown: Claude Sonnet 4.6 leads 2–1 across 4 shared unknown benchmarks (largest gap: ProofBench, 45 vs 36 — single evaluator).

2 reproduced6 unverified

A higher score does not always mean a better model. Each number is task performance under a disclosed harness. Of the 30 measurements across 15 shared benchmarks, 30 single-evaluator or undisclosed-protocol reports; 27% are independently reproduced. Check any benchmark's integrity grade before reading a win as decisive.

Frequently asked questions

Claude Sonnet 4.6 leads 9–5 across 15 shared benchmarks. Claude Opus 4.5 leads 5 benchmarks and Claude Sonnet 4.6 leads 9, with 1 tied. Higher isn't always better — see the integrity caveats on each benchmark.

Claude Opus 4.5 and Claude Sonnet 4.6 have 15 benchmarks in common in our data — those are the only rows where a direct, apples-to-apples comparison is drawn.

Claude Sonnet 4.6 leads reasoning 3–1. Claude Sonnet 4.6 leads unknown 2–1.

Follow the record

These models

Same tier

Capabilities

Reasoning

Reasoning: Claude Sonnet 4.6 leads 3–1 across 4 shared reasoning benchmarks (largest gap: SimpleQA Verified, 45.7 vs 35.5 — protocols undisclosed).

2 reproduced6 unverified

Unknown

Unknown: Claude Sonnet 4.6 leads 2–1 across 4 shared unknown benchmarks (largest gap: ProofBench, 45 vs 36 — single evaluator).

2 reproduced6 unverified

Frequently asked questions

Claude Opus 4.5 and Claude Sonnet 4.6 have 15 benchmarks in common in our data — those are the only rows where a direct, apples-to-apples comparison is drawn.

Claude Sonnet 4.6 leads reasoning 3–1. Claude Sonnet 4.6 leads unknown 2–1.