Compare · ModelsLive · 2 picked · head to head
Claude Sonnet 5 vs Grok 4.6
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Grok 4.6 wins on 11/15 benchmarks
Grok 4.6 wins 11 of 15 shared benchmarks. Leads in agentic · knowledge · coding.
Category leads
agentic·Grok 4.6knowledge·Grok 4.6coding·Grok 4.6general·Claude Sonnet 5math·Grok 4.6reasoning·Grok 4.6
Hype vs Reality
Attention vs performance
Claude Sonnet 5
#66 by perf·#10 by attention
Grok 4.6
#62 by perf·#15 by attention
Best value
Grok 4.6
1.5x better value than Claude Sonnet 5
Claude Sonnet 5
9.5 pts/$
$6.00/M
Grok 4.6
14.4 pts/$
$4.00/M
Vendor risk
Who is behind the model
Anthropic
$965.0B·Tier 1
xAI
$250.0B·Tier 1
Head to head
15 benchmarks · 2 models
Claude Sonnet 5Grok 4.6
APEX-Agents
Grok 4.6 leads by +10.8
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
Claude Sonnet 5
54.5
Grok 4.6
65.3
Chess Puzzles
Grok 4.6 leads by +5.3
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Claude Sonnet 5
31.6
Grok 4.6
36.9
Deepswe
Grok 4.6 leads by +13.6
Claude Sonnet 5
53.9
Grok 4.6
67.5
Dtbench
Grok 4.6 leads by +8.1
Claude Sonnet 5
87.5
Grok 4.6
95.5
Frontiercode
Grok 4.6 leads by +5.3
Claude Sonnet 5
42.7
Grok 4.6
48.0
FrontierMath-Tier-4-v2-Private
Grok 4.6 leads by +2.4
Claude Sonnet 5
29.3
Grok 4.6
31.7
FrontierMath-Tiers-1-3-v2-Private
Grok 4.6 leads by +0.3
Claude Sonnet 5
65.6
Grok 4.6
66.0
GPQA diamond
Grok 4.6 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 5
87.4
Grok 4.6
92.0
Lmca
Claude Sonnet 5 leads by +1.9
Claude Sonnet 5
58.9
Grok 4.6
57.0
Mystery Game Puzzles
Claude Sonnet 5 leads by +1.1
Claude Sonnet 5
28.4
Grok 4.6
27.3
OTIS Mock AIME 2024-2025
Grok 4.6 leads by +4.5
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Claude Sonnet 5
94.7
Grok 4.6
99.2
Proofbench
Claude Sonnet 5 leads by +26.0
Claude Sonnet 5
77.0
Grok 4.6
51.0
SimpleBench
Grok 4.6 leads by +18.4
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
Claude Sonnet 5
52.7
Grok 4.6
71.1
SimpleQA Verified
Grok 4.6 leads by +15.6
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
Claude Sonnet 5
33.7
Grok 4.6
49.3
WeirdML
Claude Sonnet 5 leads by +1.5
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Claude Sonnet 5
68.8
Grok 4.6
67.3
Full benchmark table
| Benchmark | Claude Sonnet 5 | Grok 4.6 |
|---|---|---|
APEX-Agents APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments. | 54.5 | 65.3 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 31.6 | 36.9 |
Deepswe | 53.9 | 67.5 |
Dtbench | 87.5 | 95.5 |
Frontiercode | 42.7 | 48.0 |
FrontierMath-Tier-4-v2-Private | 29.3 | 31.7 |
FrontierMath-Tiers-1-3-v2-Private | 65.6 | 66.0 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 87.4 | 92.0 |
Lmca | 58.9 | 57.0 |
Mystery Game Puzzles | 28.4 | 27.3 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 94.7 | 99.2 |
Proofbench | 77.0 | 51.0 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 52.7 | 71.1 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 33.7 | 49.3 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 68.8 | 67.3 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $2.00 | $10.00 | 1.0M tokens (~500 books) | $40.00 | |
| $2.00 | $6.00 | 500K tokens (~250 books) | $30.00 |
People also compared