Compare · ModelsLive · 2 picked · head to head
Claude Opus 4.6 vs Claude Sonnet 5
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Claude Sonnet 5 wins on 8/15 benchmarks
Claude Sonnet 5 wins 8 of 15 shared benchmarks. Leads in agentic · general · math.
Category leads
agentic·Claude Sonnet 5knowledge·Claude Opus 4.6general·Claude Sonnet 5coding·Claude Opus 4.6math·Claude Sonnet 5reasoning·Claude Opus 4.6
Hype vs Reality
Attention vs performance
Claude Opus 4.6
#136 by perf·#9 by attention
Claude Sonnet 5
#66 by perf·#10 by attention
Best value
Claude Sonnet 5
2.9x better value than Claude Opus 4.6
Claude Opus 4.6
3.2 pts/$
$15.00/M
Claude Sonnet 5
9.5 pts/$
$6.00/M
Vendor risk
Who is behind the model
Anthropic
$965.0B·Tier 1
Anthropic
$965.0B·Tier 1
Head to head
15 benchmarks · 2 models
Claude Opus 4.6Claude Sonnet 5
APEX-Agents
Claude Sonnet 5 leads by +8.2
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
Claude Opus 4.6
46.3
Claude Sonnet 5
54.5
Chess Puzzles
Claude Sonnet 5 leads by +18.9
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Claude Opus 4.6
12.7
Claude Sonnet 5
31.6
Dtbench
Claude Sonnet 5 leads by +2.1
Claude Opus 4.6
85.3
Claude Sonnet 5
87.5
Frontiercode
Claude Sonnet 5 leads by +16.1
Claude Opus 4.6
26.6
Claude Sonnet 5
42.7
FrontierMath-Tier-4-v2-Private
Claude Sonnet 5 leads by +2.4
Claude Opus 4.6
26.8
Claude Sonnet 5
29.3
FrontierMath-Tiers-1-3-v2-Private
Claude Opus 4.6 leads by +0.3
Claude Opus 4.6
66.0
Claude Sonnet 5
65.6
GPQA diamond
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Claude Opus 4.6
87.4
Claude Sonnet 5
87.4
GSO-Bench
Claude Opus 4.6 leads by +4.0
GSO-Bench · evaluates AI models on real-world open-source software engineering tasks, testing the ability to understand and resolve actual GitHub issues.
Claude Opus 4.6
41.2
Claude Sonnet 5
37.3
Lmca
Claude Opus 4.6 leads by +6.7
Claude Opus 4.6
65.6
Claude Sonnet 5
58.9
Mystery Game Puzzles
Claude Sonnet 5 leads by +11.0
Claude Opus 4.6
17.4
Claude Sonnet 5
28.4
OTIS Mock AIME 2024-2025
Claude Sonnet 5 leads by +0.3
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Claude Opus 4.6
94.4
Claude Sonnet 5
94.7
Proofbench
Claude Sonnet 5 leads by +27.0
Claude Opus 4.6
50.0
Claude Sonnet 5
77.0
SimpleBench
Claude Opus 4.6 leads by +8.4
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
Claude Opus 4.6
61.1
Claude Sonnet 5
52.7
SimpleQA Verified
Claude Opus 4.6 leads by +13.3
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
Claude Opus 4.6
47.0
Claude Sonnet 5
33.7
WeirdML
Claude Opus 4.6 leads by +9.2
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Claude Opus 4.6
78.0
Claude Sonnet 5
68.8
Full benchmark table
| Benchmark | Claude Opus 4.6 | Claude Sonnet 5 |
|---|---|---|
APEX-Agents APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments. | 46.3 | 54.5 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 12.7 | 31.6 |
Dtbench | 85.3 | 87.5 |
Frontiercode | 26.6 | 42.7 |
FrontierMath-Tier-4-v2-Private | 26.8 | 29.3 |
FrontierMath-Tiers-1-3-v2-Private | 66.0 | 65.6 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 87.4 | 87.4 |
GSO-Bench GSO-Bench · evaluates AI models on real-world open-source software engineering tasks, testing the ability to understand and resolve actual GitHub issues. | 41.2 | 37.3 |
Lmca | 65.6 | 58.9 |
Mystery Game Puzzles | 17.4 | 28.4 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 94.4 | 94.7 |
Proofbench | 50.0 | 77.0 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 61.1 | 52.7 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 47.0 | 33.7 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 78.0 | 68.8 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $5.00 | $25.00 | 1.0M tokens (~500 books) | $100.00 | |
| $2.00 | $10.00 | 1.0M tokens (~500 books) | $40.00 |