Compare · ModelsLive · 2 picked · head to head
Claude 3.5 Sonnet vs GPT-4o (2024-08-06)
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Claude 3.5 Sonnet wins on 9/11 benchmarks
Claude 3.5 Sonnet wins 9 of 11 shared benchmarks. Leads in coding · arena · math.
Category leads
coding·Claude 3.5 Sonnetarena·Claude 3.5 Sonnetmath·Claude 3.5 Sonnetknowledge·Claude 3.5 Sonnetreasoning·Claude 3.5 Sonnetmultimodal·GPT-4o (2024-08-06)
Hype vs Reality
Attention vs performance
Claude 3.5 Sonnet
#190 by perf·no signal
GPT-4o (2024-08-06)
#223 by perf·no signal
Best value
GPT-4o (2024-08-06)
Claude 3.5 Sonnet
n/a
no price
GPT-4o (2024-08-06)
5.7 pts/$
$6.25/M
Vendor risk
Who is behind the model
Anthropic
$965.0B·Tier 1
OpenAI
$840.0B·Tier 1
Head to head
11 benchmarks · 2 models
Claude 3.5 SonnetGPT-4o (2024-08-06)
Aider · Code Editing
Claude 3.5 Sonnet leads by +12.8
Claude 3.5 Sonnet
84.2
GPT-4o (2024-08-06)
71.4
Aider polyglot
Claude 3.5 Sonnet leads by +28.5
Aider Polyglot · measures how well AI models can edit code across multiple programming languages using the Aider coding assistant framework.
Claude 3.5 Sonnet
51.6
GPT-4o (2024-08-06)
23.1
Chatbot Arena Elo · Overall
Claude 3.5 Sonnet leads by +38.5
Claude 3.5 Sonnet
1374.0
GPT-4o (2024-08-06)
1335.5
CadEval
Claude 3.5 Sonnet leads by +22.0
CadEval · evaluates the ability to generate and reason about Computer-Aided Design code, testing spatial reasoning and engineering knowledge.
Claude 3.5 Sonnet
48.0
GPT-4o (2024-08-06)
26.0
FrontierMath-2025-02-28-Private
Claude 3.5 Sonnet leads by +1.5
FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning.
Claude 3.5 Sonnet
1.8
GPT-4o (2024-08-06)
0.3
GPQA diamond
Claude 3.5 Sonnet leads by +6.4
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Claude 3.5 Sonnet
38.7
GPT-4o (2024-08-06)
32.3
MATH level 5
GPT-4o (2024-08-06) leads by +1.6
MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics.
Claude 3.5 Sonnet
51.7
GPT-4o (2024-08-06)
53.3
MMLU
Claude 3.5 Sonnet leads by +2.9
Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge.
Claude 3.5 Sonnet
82.0
GPT-4o (2024-08-06)
79.1
OTIS Mock AIME 2024-2025
Claude 3.5 Sonnet leads by +0.1
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Claude 3.5 Sonnet
6.4
GPT-4o (2024-08-06)
6.3
SimpleBench
Claude 3.5 Sonnet leads by +11.6
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
Claude 3.5 Sonnet
13.0
GPT-4o (2024-08-06)
1.4
VideoMME
GPT-4o (2024-08-06) leads by +15.9
VideoMME · multimodal benchmark testing video understanding across diverse domains, requiring temporal reasoning and cross-frame comprehension.
Claude 3.5 Sonnet
46.7
GPT-4o (2024-08-06)
62.5
Full benchmark table
| Benchmark | Claude 3.5 Sonnet | GPT-4o (2024-08-06) |
|---|---|---|
Aider · Code Editing | 84.2 | 71.4 |
Aider polyglot Aider Polyglot · measures how well AI models can edit code across multiple programming languages using the Aider coding assistant framework. | 51.6 | 23.1 |
Chatbot Arena Elo · Overall | 1374.0 | 1335.5 |
CadEval CadEval · evaluates the ability to generate and reason about Computer-Aided Design code, testing spatial reasoning and engineering knowledge. | 48.0 | 26.0 |
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. | 1.8 | 0.3 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 38.7 | 32.3 |
MATH level 5 MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics. | 51.7 | 53.3 |
MMLU Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge. | 82.0 | 79.1 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 6.4 | 6.3 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 13.0 | 1.4 |
VideoMME VideoMME · multimodal benchmark testing video understanding across diverse domains, requiring temporal reasoning and cross-frame comprehension. | 46.7 | 62.5 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| — | — | — | — | |
| $2.50 | $10.00 | 128K tokens (~64 books) | $43.75 |