Compare · ModelsLive · 2 picked · head to head
GPT-5.2 vs Qwen3.6 Max Preview
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
GPT-5.2 wins on 8/13 benchmarks
GPT-5.2 wins 8 of 13 shared benchmarks. Leads in knowledge · general · math.
Category leads
arena·Qwen3.6 Max Previewknowledge·GPT-5.2general·GPT-5.2math·GPT-5.2reasoning·Qwen3.6 Max Previewcoding·Qwen3.6 Max Preview
Hype vs Reality
Attention vs performance
GPT-5.2
#147 by perf·#4 by attention
Qwen3.6 Max Preview
#112 by perf·#2 by attention
Best value
Qwen3.6 Max Preview
2.4x better value than GPT-5.2
GPT-5.2
6.0 pts/$
$7.88/M
Qwen3.6 Max Preview
14.2 pts/$
$3.59/M
Vendor risk
Who is behind the model
OpenAI
$840.0B·Tier 1
Alibaba (Qwen)
$293.0B·Tier 1
Head to head
13 benchmarks · 2 models
GPT-5.2Qwen3.6 Max Preview
Chatbot Arena Elo · Coding
Qwen3.6 Max Preview leads by +66.1
GPT-5.2
1415.5
Qwen3.6 Max Preview
1481.5
Chatbot Arena Elo · Overall
Qwen3.6 Max Preview leads by +24.7
GPT-5.2
1435.6
Qwen3.6 Max Preview
1460.3
Chess Puzzles
GPT-5.2 leads by +30.5
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
GPT-5.2
46.3
Qwen3.6 Max Preview
15.8
Dtbench
GPT-5.2 leads by +6.2
GPT-5.2
84.9
Qwen3.6 Max Preview
78.7
FrontierMath-2025-02-28-Private
GPT-5.2 leads by +0.2
FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning.
GPT-5.2
40.7
Qwen3.6 Max Preview
40.5
FrontierMath-Tier-4-2025-07-01-Private
GPT-5.2 leads by +11.9
FrontierMath Tier 4 (Jul 2025) · the most challenging tier of frontier mathematics, containing problems that push the absolute limits of AI mathematical reasoning.
GPT-5.2
18.8
Qwen3.6 Max Preview
6.9
GPQA diamond
GPT-5.2 leads by +5.4
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
GPT-5.2
88.5
Qwen3.6 Max Preview
83.2
Lmca
GPT-5.2 leads by +1.6
GPT-5.2
51.7
Qwen3.6 Max Preview
50.0
Mystery Game Puzzles
GPT-5.2 leads by +4.4
GPT-5.2
15.2
Qwen3.6 Max Preview
10.8
OTIS Mock AIME 2024-2025
GPT-5.2 leads by +5.0
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
GPT-5.2
96.1
Qwen3.6 Max Preview
91.1
SimpleBench
Qwen3.6 Max Preview leads by +20.6
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
GPT-5.2
35.0
Qwen3.6 Max Preview
55.6
SimpleQA Verified
Qwen3.6 Max Preview leads by +14.9
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
GPT-5.2
37.1
Qwen3.6 Max Preview
52.0
SWE-Bench verified
Qwen3.6 Max Preview leads by +2.9
SWE-bench Verified · 500 human-validated tasks from 12 real Python repositories (Django, Flask, scikit-learn, sympy, and others). Each task requires the model to produce a git patch that resolves a real GitHub issue and passes the test suite. The verified subset eliminates ambiguous tasks from the original SWE-bench. Claude Mythos Preview leads at 93.9%, crossing 90% for the first time in 2026. Opus 4.6 scores 80.8%. The benchmark remains the most-cited evaluation for code-generation capability.
GPT-5.2
73.8
Qwen3.6 Max Preview
76.7
Full benchmark table
| Benchmark | GPT-5.2 | Qwen3.6 Max Preview |
|---|---|---|
Chatbot Arena Elo · Coding | 1415.5 | 1481.5 |
Chatbot Arena Elo · Overall | 1435.6 | 1460.3 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 46.3 | 15.8 |
Dtbench | 84.9 | 78.7 |
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. | 40.7 | 40.5 |
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. | 18.8 | 6.9 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 88.5 | 83.2 |
Lmca | 51.7 | 50.0 |
Mystery Game Puzzles | 15.2 | 10.8 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 96.1 | 91.1 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 35.0 | 55.6 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 37.1 | 52.0 |
SWE-Bench verified SWE-bench Verified · 500 human-validated tasks from 12 real Python repositories (Django, Flask, scikit-learn, sympy, and others). Each task requires the model to produce a git patch that resolves a real GitHub issue and passes the test suite. The verified subset eliminates ambiguous tasks from the original SWE-bench. Claude Mythos Preview leads at 93.9%, crossing 90% for the first time in 2026. Opus 4.6 scores 80.8%. The benchmark remains the most-cited evaluation for code-generation capability. | 73.8 | 76.7 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $1.75 | $14.00 | 400K tokens (~200 books) | $48.13 | |
| $1.03 | $6.16 | 262K tokens (~131 books) | $23.11 |
People also compared