Compare · ModelsLive · 2 picked · head to head
o3 Pro vs o3
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
o3 Pro wins on 6/8 benchmarks
o3 Pro wins 6 of 8 shared benchmarks. Leads in coding · knowledge.
Category leads
coding·o3 Proreasoning·o3knowledge·o3 Pro
Hype vs Reality
Attention vs performance
o3 Pro
#33 by perf·no signal
o3
#67 by perf·no signal
Best value
o3
9.0x better value than o3 Pro
o3 Pro
1.2 pts/$
$50.00/M
o3
11.0 pts/$
$5.00/M
Vendor risk
Who is behind the model
OpenAI
$840.0B·Tier 1
OpenAI
$840.0B·Tier 1
Head to head
8 benchmarks · 2 models
o3 Proo3
Aider polyglot
o3 Pro leads by +3.6
Aider Polyglot · measures how well AI models can edit code across multiple programming languages using the Aider coding assistant framework.
o3 Pro
84.9
o3
81.3
ARC-AGI
o3 leads by +1.5
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
o3 Pro
59.3
o3
60.8
ARC-AGI-2
o3 leads by +1.7
ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data.
o3 Pro
4.9
o3
6.5
Fiction.LiveBench
o3 Pro leads by +8.3
Fiction.LiveBench · a continuously updated benchmark using recently published fiction to test reading comprehension and reasoning, preventing data contamination.
o3 Pro
97.2
o3
88.9
Lech Mazur Writing
o3 Pro leads by +2.4
Lech Mazur Writing · evaluates creative writing ability, assessing prose quality, narrative coherence, and stylistic sophistication.
o3 Pro
86.3
o3
83.9
Professional Reasoning · Finance
o3 Pro leads by +1.4
o3 Pro
49.1
o3
47.7
Professional Reasoning · Legal
o3 Pro leads by +1.1
o3 Pro
49.7
o3
48.6
WeirdML
o3 Pro leads by +5.8
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
o3 Pro
58.2
o3
52.4
Full benchmark table
| Benchmark | o3 Pro | o3 |
|---|---|---|
Aider polyglot Aider Polyglot · measures how well AI models can edit code across multiple programming languages using the Aider coding assistant framework. | 84.9 | 81.3 |
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 59.3 | 60.8 |
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. | 4.9 | 6.5 |
Fiction.LiveBench Fiction.LiveBench · a continuously updated benchmark using recently published fiction to test reading comprehension and reasoning, preventing data contamination. | 97.2 | 88.9 |
Lech Mazur Writing Lech Mazur Writing · evaluates creative writing ability, assessing prose quality, narrative coherence, and stylistic sophistication. | 86.3 | 83.9 |
Professional Reasoning · Finance | 49.1 | 47.7 |
Professional Reasoning · Legal | 49.7 | 48.6 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 58.2 | 52.4 |
Pricing · per 1M tokens · projected $/mo at 10M tokens