Compare · ModelsLive · 2 picked · head to head
Claude Sonnet 4 vs Grok 3
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Claude Sonnet 4 wins on 11/13 benchmarks
Claude Sonnet 4 wins 11 of 13 shared benchmarks. Leads in coding · agentic · reasoning.
Category leads
coding·Claude Sonnet 4agentic·Claude Sonnet 4reasoning·Claude Sonnet 4knowledge·Claude Sonnet 4math·Claude Sonnet 4
Hype vs Reality
Attention vs performance
Claude Sonnet 4
#157 by perf·#10 by attention
Grok 3
#155 by perf·no signal
Vendor risk
Who is behind the model
Anthropic
$965.0B·Tier 1
xAI
$250.0B·Tier 1
Head to head
13 benchmarks · 2 models
Claude Sonnet 4Grok 3
Aider polyglot
Claude Sonnet 4 leads by +8.0
Aider Polyglot · measures how well AI models can edit code across multiple programming languages using the Aider coding assistant framework.
Claude Sonnet 4
61.3
Grok 3
53.3
APEX-Agents
Claude Sonnet 4 leads by +7.2
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
Claude Sonnet 4
9.3
Grok 3
2.1
ARC-AGI
Claude Sonnet 4 leads by +34.5
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
Claude Sonnet 4
40.0
Grok 3
5.5
ARC-AGI-2
Claude Sonnet 4 leads by +5.9
ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data.
Claude Sonnet 4
5.9
Grok 3
0.0
Fiction.LiveBench
Grok 3 leads by +11.4
Fiction.LiveBench · a continuously updated benchmark using recently published fiction to test reading comprehension and reasoning, preventing data contamination.
Claude Sonnet 4
46.9
Grok 3
58.3
FrontierMath-2025-02-28-Private
Claude Sonnet 4 leads by +0.6
FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning.
Claude Sonnet 4
7.3
Grok 3
6.7
FrontierMath-Tier-4-2025-07-01-Private
FrontierMath Tier 4 (Jul 2025) · the most challenging tier of frontier mathematics, containing problems that push the absolute limits of AI mathematical reasoning.
Claude Sonnet 4
0.0
Grok 3
0.0
GPQA diamond
Claude Sonnet 4 leads by +4.6
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Claude Sonnet 4
72.3
Grok 3
67.7
Lech Mazur Writing
Claude Sonnet 4 leads by +5.0
Lech Mazur Writing · evaluates creative writing ability, assessing prose quality, narrative coherence, and stylistic sophistication.
Claude Sonnet 4
81.4
Grok 3
76.4
MATH level 5
Grok 3 leads by +4.4
MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics.
Claude Sonnet 4
84.4
Grok 3
88.8
OTIS Mock AIME 2024-2025
Claude Sonnet 4 leads by +15.6
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Claude Sonnet 4
71.1
Grok 3
55.5
SimpleBench
Claude Sonnet 4 leads by +11.3
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
Claude Sonnet 4
34.6
Grok 3
23.3
WeirdML
Claude Sonnet 4 leads by +8.9
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Claude Sonnet 4
46.1
Grok 3
37.2
Full benchmark table
| Benchmark | Claude Sonnet 4 | Grok 3 |
|---|---|---|
Aider polyglot Aider Polyglot · measures how well AI models can edit code across multiple programming languages using the Aider coding assistant framework. | 61.3 | 53.3 |
APEX-Agents APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments. | 9.3 | 2.1 |
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 40.0 | 5.5 |
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. | 5.9 | 0.0 |
Fiction.LiveBench Fiction.LiveBench · a continuously updated benchmark using recently published fiction to test reading comprehension and reasoning, preventing data contamination. | 46.9 | 58.3 |
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. | 7.3 | 6.7 |
FrontierMath-Tier-4-2025-07-01-Private FrontierMath Tier 4 (Jul 2025) · the most challenging tier of frontier mathematics, containing problems that push the absolute limits of AI mathematical reasoning. | 0.0 | 0.0 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 72.3 | 67.7 |
Lech Mazur Writing Lech Mazur Writing · evaluates creative writing ability, assessing prose quality, narrative coherence, and stylistic sophistication. | 81.4 | 76.4 |
MATH level 5 MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics. | 84.4 | 88.8 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 71.1 | 55.5 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 34.6 | 23.3 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 46.1 | 37.2 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $3.00 | $15.00 | 200K tokens (~100 books) | $60.00 | |
| — | — | — | — |