Compare · ModelsLive · 2 picked · head to head
o1-mini vs Claude 3 Sonnet
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
o1-mini wins on 4/4 benchmarks
o1-mini wins 4 of 4 shared benchmarks. Leads in knowledge · math · coding.
Category leads
knowledge·o1-minimath·o1-minicoding·o1-mini
Hype vs Reality
Attention vs performance
o1-mini
#171 by perf·no signal
Claude 3 Sonnet
#192 by perf·no signal
Vendor risk
Who is behind the model
OpenAI
$840.0B·Tier 1
Anthropic
$380.0B·Tier 1
Head to head
4 benchmarks · 2 models
o1-miniClaude 3 Sonnet
GPQA diamond
o1-mini leads by +29.0
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
o1-mini
49.8
Claude 3 Sonnet
20.8
MATH level 5
o1-mini leads by +71.0
MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics.
o1-mini
89.2
Claude 3 Sonnet
18.2
OTIS Mock AIME 2024-2025
o1-mini leads by +44.5
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
o1-mini
46.9
Claude 3 Sonnet
2.4
WeirdML
o1-mini leads by +26.2
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
o1-mini
36.3
Claude 3 Sonnet
10.2
Full benchmark table
| Benchmark | o1-mini | Claude 3 Sonnet |
|---|---|---|
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 49.8 | 20.8 |
MATH level 5 MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics. | 89.2 | 18.2 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 46.9 | 2.4 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 36.3 | 10.2 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| — | — | — | — | |
| — | — | — | — |