Compare · ModelsLive · 2 picked · head to head
Claude Opus 4.6 vs o3 Pro
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Claude Opus 4.6 wins on 7/7 benchmarks
Claude Opus 4.6 wins 7 of 7 shared benchmarks. Leads in reasoning · general · knowledge.
Category leads
reasoning·Claude Opus 4.6general·Claude Opus 4.6knowledge·Claude Opus 4.6coding·Claude Opus 4.6
Hype vs Reality
Attention vs performance
Claude Opus 4.6
#136 by perf·#9 by attention
o3 Pro
#42 by perf·no signal
Best value
Claude Opus 4.6
2.6x better value than o3 Pro
Claude Opus 4.6
3.2 pts/$
$15.00/M
o3 Pro
1.2 pts/$
$50.00/M
Vendor risk
Who is behind the model
Anthropic
$965.0B·Tier 1
OpenAI
$840.0B·Tier 1
Head to head
7 benchmarks · 2 models
Claude Opus 4.6o3 Pro
ARC-AGI
Claude Opus 4.6 leads by +34.7
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
Claude Opus 4.6
94.0
o3 Pro
59.3
ARC-AGI-2
Claude Opus 4.6 leads by +64.3
ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data.
Claude Opus 4.6
69.2
o3 Pro
4.9
Dtbench
Claude Opus 4.6 leads by +7.1
Claude Opus 4.6
85.3
o3 Pro
78.2
Lmca
Claude Opus 4.6 leads by +20.3
Claude Opus 4.6
65.6
o3 Pro
45.3
Professional Reasoning · Finance
Claude Opus 4.6 leads by +4.2
Claude Opus 4.6
53.3
o3 Pro
49.1
Professional Reasoning · Legal
Claude Opus 4.6 leads by +2.6
Claude Opus 4.6
52.3
o3 Pro
49.7
WeirdML
Claude Opus 4.6 leads by +19.7
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Claude Opus 4.6
78.0
o3 Pro
58.2
Full benchmark table
| Benchmark | Claude Opus 4.6 | o3 Pro |
|---|---|---|
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 94.0 | 59.3 |
ARC-AGI-2 ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data. | 69.2 | 4.9 |
Dtbench | 85.3 | 78.2 |
Lmca | 65.6 | 45.3 |
Professional Reasoning · Finance | 53.3 | 49.1 |
Professional Reasoning · Legal | 52.3 | 49.7 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 78.0 | 58.2 |
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 | |
| $20.00 | $80.00 | 200K tokens (~100 books) | $350.00 |