Compare · ModelsLive · 2 picked · head to head
GPT-5 Mini vs o3
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
GPT-5 Mini wins on 10/19 benchmarks
GPT-5 Mini wins 10 of 19 shared benchmarks. Leads in math · language · coding.
Category leads
reasoning·o3knowledge·o3math·GPT-5 Minilanguage·GPT-5 Minicoding·GPT-5 Mini
Hype vs Reality
Attention vs performance
GPT-5 Mini
#63 by perf·no signal
o3
#67 by perf·no signal
Best value
GPT-5 Mini
4.5x better value than o3
GPT-5 Mini
49.8 pts/$
$1.13/M
o3
11.0 pts/$
$5.00/M
Vendor risk
Who is behind the model
OpenAI
$840.0B·Tier 1
OpenAI
$840.0B·Tier 1
Head to head
19 benchmarks · 2 models
GPT-5 Minio3
ARC-AGI
o3 leads by +6.5
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
GPT-5 Mini
54.3
o3
60.8
ARC-AGI-2
o3 leads by +2.1
ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data.
GPT-5 Mini
4.4
o3
6.5
Fiction.LiveBench
o3 leads by +19.5
Fiction.LiveBench · a continuously updated benchmark using recently published fiction to test reading comprehension and reasoning, preventing data contamination.
GPT-5 Mini
69.4
o3
88.9
FrontierMath-2025-02-28-Private
GPT-5 Mini leads by +8.5
FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning.
GPT-5 Mini
27.2
o3
18.7
FrontierMath-Tier-4-2025-07-01-Private
GPT-5 Mini leads by +4.2
FrontierMath Tier 4 (Jul 2025) · the most challenging tier of frontier mathematics, containing problems that push the absolute limits of AI mathematical reasoning.
GPT-5 Mini
6.3
o3
2.1
GPQA diamond
o3 leads by +9.1
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
GPT-5 Mini
66.7
o3
75.8
HELM · GPQA
GPT-5 Mini leads by +0.3
GPT-5 Mini
75.6
o3
75.3
HELM · IFEval
GPT-5 Mini leads by +5.8
GPT-5 Mini
92.7
o3
86.9
HELM · MMLU-Pro
o3 leads by +2.4
GPT-5 Mini
83.5
o3
85.9
HELM · Omni-MATH
GPT-5 Mini leads by +0.8
GPT-5 Mini
72.2
o3
71.4
HELM · WildBench
o3 leads by +0.6
GPT-5 Mini
85.5
o3
86.1
HLE
o3 leads by +0.9
HLE (Humanity's Last Exam) · crowdsourced expert-level questions designed to be among the hardest possible challenges for AI systems across all domains.
GPT-5 Mini
15.4
o3
16.3
MATH level 5
GPT-5 Mini leads by +0.1
MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics.
GPT-5 Mini
97.8
o3
97.8
OTIS Mock AIME 2024-2025
GPT-5 Mini leads by +2.8
OTIS Mock AIME 2024–2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
GPT-5 Mini
86.7
o3
83.9
SimpleQA Verified
o3 leads by +32.0
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
GPT-5 Mini
21.0
o3
53.0
SWE-Bench verified
GPT-5 Mini leads by +2.4
GPT-5 Mini
64.7
o3
62.3
SWE-Bench Verified (Bash Only)
GPT-5 Mini leads by +1.4
SWE-Bench Verified (Bash Only) · a curated subset of SWE-bench where models fix real Python repository bugs using only bash commands, no agent frameworks.
GPT-5 Mini
59.8
o3
58.4
VPCT
o3 leads by +17.7
VPCT (Visual Pattern Completion Test) · tests visual reasoning and pattern recognition by having models complete visual sequences and transformations.
GPT-5 Mini
10.3
o3
28.0
WeirdML
GPT-5 Mini leads by +0.3
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
GPT-5 Mini
52.7
o3
52.4
Full benchmark table
| Benchmark | GPT-5 Mini | o3 |
|---|---|---|
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 54.3 | 60.8 |
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. | 4.4 | 6.5 |
Fiction.LiveBench Fiction.LiveBench · a continuously updated benchmark using recently published fiction to test reading comprehension and reasoning, preventing data contamination. | 69.4 | 88.9 |
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. | 27.2 | 18.7 |
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. | 6.3 | 2.1 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 66.7 | 75.8 |
HELM · GPQA | 75.6 | 75.3 |
HELM · IFEval | 92.7 | 86.9 |
HELM · MMLU-Pro | 83.5 | 85.9 |
HELM · Omni-MATH | 72.2 | 71.4 |
HELM · WildBench | 85.5 | 86.1 |
HLE HLE (Humanity's Last Exam) · crowdsourced expert-level questions designed to be among the hardest possible challenges for AI systems across all domains. | 15.4 | 16.3 |
MATH level 5 MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics. | 97.8 | 97.8 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024–2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 86.7 | 83.9 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 21.0 | 53.0 |
SWE-Bench verified | 64.7 | 62.3 |
SWE-Bench Verified (Bash Only) SWE-Bench Verified (Bash Only) · a curated subset of SWE-bench where models fix real Python repository bugs using only bash commands, no agent frameworks. | 59.8 | 58.4 |
VPCT VPCT (Visual Pattern Completion Test) · tests visual reasoning and pattern recognition by having models complete visual sequences and transformations. | 10.3 | 28.0 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 52.7 | 52.4 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| GPT-5 Mini | $0.25 | $2.00 | 400K tokens (~200 books) | $6.88 |
| o3 | $2.00 | $8.00 | 200K tokens (~100 books) | $35.00 |