Compare · ModelsLive · 2 picked · head to head
Claude Opus 4 vs o1-preview
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Claude Opus 4 wins on 6/7 benchmarks
Claude Opus 4 wins 6 of 7 shared benchmarks. Leads in reasoning · coding · knowledge.
Category leads
reasoning·Claude Opus 4coding·Claude Opus 4knowledge·Claude Opus 4math·Claude Opus 4
Hype vs Reality
Attention vs performance
Claude Opus 4
#133 by perf·no signal
o1-preview
#136 by perf·no signal
Vendor risk
Who is behind the model
Anthropic
$380.0B·Tier 1
OpenAI
$840.0B·Tier 1
Head to head
7 benchmarks · 2 models
Claude Opus 4o1-preview
ARC-AGI
Claude Opus 4 leads by +17.7
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
Claude Opus 4
35.7
o1-preview
18.0
Cybench
Claude Opus 4 leads by +28.0
Cybench · evaluates AI on real Capture-The-Flag cybersecurity challenges, testing vulnerability analysis, exploitation, and security reasoning.
Claude Opus 4
38.0
o1-preview
10.0
GPQA diamond
Claude Opus 4 leads by +34.6
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
68.3
o1-preview
33.8
MATH level 5
Claude Opus 4 leads by +3.4
MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics.
Claude Opus 4
85.0
o1-preview
81.7
OTIS Mock AIME 2024-2025
Claude Opus 4 leads by +33.4
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Claude Opus 4
64.4
o1-preview
31.0
SimpleBench
Claude Opus 4 leads by +20.5
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
Claude Opus 4
50.6
o1-preview
30.0
WeirdML
o1-preview leads by +4.2
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Claude Opus 4
43.4
o1-preview
47.6
Full benchmark table
| Benchmark | Claude Opus 4 | o1-preview |
|---|---|---|
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 35.7 | 18.0 |
Cybench Cybench · evaluates AI on real Capture-The-Flag cybersecurity challenges, testing vulnerability analysis, exploitation, and security reasoning. | 38.0 | 10.0 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 68.3 | 33.8 |
MATH level 5 MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics. | 85.0 | 81.7 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 64.4 | 31.0 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 50.6 | 30.0 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 43.4 | 47.6 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $15.00 | $75.00 | 200K tokens (~100 books) | $300.00 | |
| — | — | — | — |
People also compared