Compare · ModelsLive · 3 picked · head to head
GPT-5 vs o3 Pro vs Qwen3 Max
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
GPT-5 wins on 14/17 benchmarks
GPT-5 wins 14 of 17 shared benchmarks. Leads in general · knowledge · coding.
Category leads
general·GPT-5knowledge·GPT-5coding·GPT-5reasoning·GPT-5math·GPT-5
Hype vs Reality
Attention vs performance
GPT-5
#115 by perf·#9 by attention
o3 Pro
#49 by perf·no signal
Qwen3 Max
#162 by perf·#1 by attention
Best value
Qwen3 Max
2.1x better value than GPT-5
GPT-5
9.3 pts/$
$5.63/M
o3 Pro
1.2 pts/$
$50.00/M
Qwen3 Max
19.9 pts/$
$2.34/M
Vendor risk
Who is behind the model
OpenAI
$840.0B·Tier 1
OpenAI
$840.0B·Tier 1
Alibaba (Qwen)
$293.0B·Tier 1
Head to head
17 benchmarks · 3 models
GPT-5o3 ProQwen3 Max
Dtbench
GPT-5 leads by +6.2
GPT-5
84.5
o3 Pro
78.2
Qwen3 Max
70.2
Fiction.LiveBench
Fiction.LiveBench · a continuously updated benchmark using recently published fiction to test reading comprehension and reasoning, preventing data contamination.
GPT-5
97.2
o3 Pro
97.2
Qwen3 Max
66.7
Lech Mazur Writing
Qwen3 Max leads by +0.8
Lech Mazur Writing · evaluates creative writing ability, assessing prose quality, narrative coherence, and stylistic sophistication.
GPT-5
86.0
o3 Pro
86.3
Qwen3 Max
87.1
Lmca
GPT-5 leads by +1.8
GPT-5
47.0
o3 Pro
45.3
Qwen3 Max
33.3
Aider polyglot
GPT-5 leads by +3.1
Aider Polyglot · measures how well AI models can edit code across multiple programming languages using the Aider coding assistant framework.
GPT-5
88.0
o3 Pro
84.9
ARC-AGI
GPT-5 leads by +6.4
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
GPT-5
65.7
o3 Pro
59.3
ARC-AGI-2
GPT-5 leads by +5.0
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
9.9
o3 Pro
4.9
Chess Puzzles
GPT-5 leads by +33.0
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
GPT-5
37.0
Qwen3 Max
4.0
FrontierMath-Tiers-1-3-v2-Private
GPT-5 leads by +38.4
GPT-5
58.4
Qwen3 Max
19.9
GPQA diamond
GPT-5 leads by +21.5
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
GPT-5
85.0
Qwen3 Max
63.5
MATH level 5
GPT-5 leads by +1.0
MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics.
GPT-5
98.1
Qwen3 Max
97.1
Mystery Game Puzzles
GPT-5 leads by +15.2
GPT-5
15.2
Qwen3 Max
0.0
OTIS Mock AIME 2024-2025
GPT-5 leads by +18.1
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
GPT-5
91.4
Qwen3 Max
73.3
Professional Reasoning · Finance
GPT-5 leads by +2.2
GPT-5
51.3
o3 Pro
49.1
Professional Reasoning · Legal
o3 Pro leads by +0.7
GPT-5
49.0
o3 Pro
49.7
SimpleQA Verified
Qwen3 Max leads by +16.9
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
GPT-5
50.6
Qwen3 Max
67.5
WeirdML
GPT-5 leads by +2.5
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
GPT-5
60.7
o3 Pro
58.2
Full benchmark table
| Benchmark | GPT-5 | o3 Pro | Qwen3 Max |
|---|---|---|---|
Dtbench | 84.5 | 78.2 | 70.2 |
Fiction.LiveBench Fiction.LiveBench · a continuously updated benchmark using recently published fiction to test reading comprehension and reasoning, preventing data contamination. | 97.2 | 97.2 | 66.7 |
Lech Mazur Writing Lech Mazur Writing · evaluates creative writing ability, assessing prose quality, narrative coherence, and stylistic sophistication. | 86.0 | 86.3 | 87.1 |
Lmca | 47.0 | 45.3 | 33.3 |
Aider polyglot Aider Polyglot · measures how well AI models can edit code across multiple programming languages using the Aider coding assistant framework. | 88.0 | 84.9 | n/a |
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 65.7 | 59.3 | n/a |
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. | 9.9 | 4.9 | n/a |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 37.0 | n/a | 4.0 |
FrontierMath-Tiers-1-3-v2-Private | 58.4 | n/a | 19.9 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 85.0 | n/a | 63.5 |
MATH level 5 MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics. | 98.1 | n/a | 97.1 |
Mystery Game Puzzles | 15.2 | n/a | 0.0 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 91.4 | n/a | 73.3 |
Professional Reasoning · Finance | 51.3 | 49.1 | n/a |
Professional Reasoning · Legal | 49.0 | 49.7 | n/a |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 50.6 | n/a | 67.5 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 60.7 | 58.2 | n/a |
Pricing · per 1M tokens · projected $/mo at 10M tokens