Compare · ModelsLive · 2 picked · head to head
o1 vs Claude 3.7 Sonnet
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
o1 wins on 7/12 benchmarks
o1 wins 7 of 12 shared benchmarks. Leads in reasoning · math.
Category leads
coding·Claude 3.7 Sonnetreasoning·o1knowledge·Claude 3.7 Sonnetmath·o1
Hype vs Reality
Attention vs performance
o1
#59 by perf·no signal
Claude 3.7 Sonnet
#103 by perf·no signal
Best value
Claude 3.7 Sonnet
3.5x better value than o1
o1
1.5 pts/$
$37.50/M
Claude 3.7 Sonnet
5.3 pts/$
$9.00/M
Vendor risk
Who is behind the model
OpenAI
$840.0B·Tier 1
Anthropic
$380.0B·Tier 1
Head to head
12 benchmarks · 2 models
o1Claude 3.7 Sonnet
Aider polyglot
Claude 3.7 Sonnet leads by +3.2
Aider Polyglot · measures how well AI models can edit code across multiple programming languages using the Aider coding assistant framework.
o1
61.7
Claude 3.7 Sonnet
64.9
ARC-AGI
o1 leads by +2.1
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
o1
30.7
Claude 3.7 Sonnet
28.6
CadEval
o1 leads by +2.0
CadEval · evaluates the ability to generate and reason about Computer-Aided Design code, testing spatial reasoning and engineering knowledge.
o1
56.0
Claude 3.7 Sonnet
54.0
Fiction.LiveBench
Fiction.LiveBench · a continuously updated benchmark using recently published fiction to test reading comprehension and reasoning, preventing data contamination.
o1
83.3
Claude 3.7 Sonnet
83.3
FrontierMath-2025-02-28-Private
o1 leads by +5.2
FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning.
o1
9.3
Claude 3.7 Sonnet
4.1
GeoBench
o1 leads by +12.0
GeoBench · tests geographic knowledge and spatial reasoning across countries, landmarks, coordinates, and geopolitical understanding.
o1
80.0
Claude 3.7 Sonnet
68.0
GPQA diamond
Claude 3.7 Sonnet leads by +4.0
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
o1
69.0
Claude 3.7 Sonnet
73.0
Lech Mazur Writing
Claude 3.7 Sonnet leads by +10.9
Lech Mazur Writing · evaluates creative writing ability, assessing prose quality, narrative coherence, and stylistic sophistication.
o1
70.2
Claude 3.7 Sonnet
81.1
MATH level 5
o1 leads by +3.5
MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics.
o1
94.7
Claude 3.7 Sonnet
91.2
OTIS Mock AIME 2024-2025
o1 leads by +15.6
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
o1
73.3
Claude 3.7 Sonnet
57.7
SimpleBench
Claude 3.7 Sonnet leads by +7.6
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
o1
28.1
Claude 3.7 Sonnet
35.7
VPCT
Claude 3.7 Sonnet leads by +3.0
VPCT (Visual Pattern Completion Test) · tests visual reasoning and pattern recognition by having models complete visual sequences and transformations.
o1
5.5
Claude 3.7 Sonnet
8.5
Full benchmark table
| Benchmark | o1 | Claude 3.7 Sonnet |
|---|---|---|
Aider polyglot Aider Polyglot · measures how well AI models can edit code across multiple programming languages using the Aider coding assistant framework. | 61.7 | 64.9 |
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 30.7 | 28.6 |
CadEval CadEval · evaluates the ability to generate and reason about Computer-Aided Design code, testing spatial reasoning and engineering knowledge. | 56.0 | 54.0 |
Fiction.LiveBench Fiction.LiveBench · a continuously updated benchmark using recently published fiction to test reading comprehension and reasoning, preventing data contamination. | 83.3 | 83.3 |
FrontierMath-2025-02-28-Private FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning. | 9.3 | 4.1 |
GeoBench GeoBench · tests geographic knowledge and spatial reasoning across countries, landmarks, coordinates, and geopolitical understanding. | 80.0 | 68.0 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 69.0 | 73.0 |
Lech Mazur Writing Lech Mazur Writing · evaluates creative writing ability, assessing prose quality, narrative coherence, and stylistic sophistication. | 70.2 | 81.1 |
MATH level 5 MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics. | 94.7 | 91.2 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 73.3 | 57.7 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 28.1 | 35.7 |
VPCT VPCT (Visual Pattern Completion Test) · tests visual reasoning and pattern recognition by having models complete visual sequences and transformations. | 5.5 | 8.5 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $15.00 | $60.00 | 200K tokens (~100 books) | $262.50 | |
| $3.00 | $15.00 | 200K tokens (~100 books) | $60.00 |