Compare · ModelsLive · 2 picked · head to head
Grok 4.20 vs Grok 4.7
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Grok 4.7 wins on 9/11 benchmarks
Grok 4.7 wins 9 of 11 shared benchmarks. Leads in reasoning · knowledge · general.
Category leads
reasoning·Grok 4.7knowledge·Grok 4.7general·Grok 4.7math·Grok 4.7
Hype vs Reality
Attention vs performance
Grok 4.20
#136 by perf·#18 by attention
Grok 4.7
#107 by perf·#18 by attention
Best value
Grok 4.20
2.0x better value than Grok 4.7
Grok 4.20
26.2 pts/$
$1.88/M
Grok 4.7
13.2 pts/$
$4.00/M
Vendor risk
Who is behind the model
xAI
$250.0B·Tier 1
xAI
$250.0B·Tier 1
Head to head
11 benchmarks · 2 models
Grok 4.20Grok 4.7
ARC-AGI
Grok 4.7 leads by +0.7
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
Grok 4.20
89.5
Grok 4.7
90.2
ARC-AGI-2
Grok 4.20 leads by +3.8
ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data.
Grok 4.20
65.1
Grok 4.7
61.4
Chess Puzzles
Grok 4.7 leads by +12.6
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Grok 4.20
25.3
Grok 4.7
37.9
Dtbench
Grok 4.7 leads by +9.8
Grok 4.20
83.5
Grok 4.7
93.3
FrontierMath-Tier-4-v2-Private
Grok 4.20
17.1
Grok 4.7
17.1
FrontierMath-Tiers-1-3-v2-Private
Grok 4.7 leads by +8.1
Grok 4.20
44.9
Grok 4.7
53.0
GPQA diamond
Grok 4.7 leads by +4.5
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Grok 4.20
85.8
Grok 4.7
90.3
Lmca
Grok 4.7 leads by +12.6
Grok 4.20
45.5
Grok 4.7
58.1
OTIS Mock AIME 2024-2025
Grok 4.7 leads by +5.8
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Grok 4.20
92.2
Grok 4.7
98.0
Proofbench
Grok 4.7 leads by +20.0
Grok 4.20
14.0
Grok 4.7
34.0
SimpleQA Verified
Grok 4.7 leads by +25.8
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
Grok 4.20
30.2
Grok 4.7
56.0
Full benchmark table
| Benchmark | Grok 4.20 | Grok 4.7 |
|---|---|---|
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 89.5 | 90.2 |
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. | 65.1 | 61.4 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 25.3 | 37.9 |
Dtbench | 83.5 | 93.3 |
FrontierMath-Tier-4-v2-Private | 17.1 | 17.1 |
FrontierMath-Tiers-1-3-v2-Private | 44.9 | 53.0 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 85.8 | 90.3 |
Lmca | 45.5 | 58.1 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 92.2 | 98.0 |
Proofbench | 14.0 | 34.0 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 30.2 | 56.0 |