Compare · ModelsLive · 2 picked · head to head
Grok 3 Mini vs o1-preview
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Grok 3 Mini wins on 3/5 benchmarks
Grok 3 Mini wins 3 of 5 shared benchmarks. Leads in knowledge · math.
Category leads
reasoning·o1-previewknowledge·Grok 3 Minimath·Grok 3 Minicoding·o1-preview
Hype vs Reality
Attention vs performance
Grok 3 Mini
#110 by perf·no signal
o1-preview
#136 by perf·no signal
Vendor risk
Who is behind the model
xAI
$250.0B·Tier 1
OpenAI
$840.0B·Tier 1
Head to head
5 benchmarks · 2 models
Grok 3 Minio1-preview
ARC-AGI
o1-preview leads by +1.5
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
Grok 3 Mini
16.5
o1-preview
18.0
GPQA diamond
Grok 3 Mini leads by +34.6
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Grok 3 Mini
68.3
o1-preview
33.8
MATH level 5
Grok 3 Mini leads by +9.3
MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics.
Grok 3 Mini
90.9
o1-preview
81.7
OTIS Mock AIME 2024-2025
Grok 3 Mini leads by +46.7
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Grok 3 Mini
77.8
o1-preview
31.0
WeirdML
o1-preview leads by +5.0
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Grok 3 Mini
42.6
o1-preview
47.6
Full benchmark table
| Benchmark | Grok 3 Mini | o1-preview |
|---|---|---|
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 16.5 | 18.0 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 68.3 | 33.8 |
MATH level 5 MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics. | 90.9 | 81.7 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 77.8 | 31.0 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 42.6 | 47.6 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $0.30 | $0.50 | 131K tokens (~66 books) | $3.50 | |
| — | — | — | — |
People also compared