Compare · ModelsLive · 2 picked · head to head
GPT-5.1 vs Qwen3 235B A22B Thinking 2507
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
GPT-5.1 wins on 11/11 benchmarks
GPT-5.1 wins 11 of 11 shared benchmarks. Leads in arena · knowledge · general.
Category leads
arena·GPT-5.1knowledge·GPT-5.1general·GPT-5.1math·GPT-5.1coding·GPT-5.1
Hype vs Reality
Attention vs performance
GPT-5.1
#156 by perf·#9 by attention
Qwen3 235B A22B Thinking 2507
#108 by perf·#1 by attention
Best value
Qwen3 235B A22B Thinking 2507
3.2x better value than GPT-5.1
GPT-5.1
8.4 pts/$
$5.63/M
Qwen3 235B A22B Thinking 2507
26.9 pts/$
$1.98/M
Vendor risk
Who is behind the model
OpenAI
$840.0B·Tier 1
Alibaba (Qwen)
$293.0B·Tier 1
Head to head
11 benchmarks · 2 models
GPT-5.1Qwen3 235B A22B Thinking 2507
Chatbot Arena Elo · Overall
GPT-5.1 leads by +38.5
GPT-5.1
1438.5
Qwen3 235B A22B Thinking 2507
1400.0
Chess Puzzles
GPT-5.1 leads by +31.6
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
GPT-5.1
39.0
Qwen3 235B A22B Thinking 2507
7.4
Dtbench
GPT-5.1 leads by +16.4
GPT-5.1
83.5
Qwen3 235B A22B Thinking 2507
67.1
FrontierMath-2025-02-28-Private
GPT-5.1 leads by +16.1
FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning.
GPT-5.1
31.0
Qwen3 235B A22B Thinking 2507
14.9
FrontierMath-Tier-4-2025-07-01-Private
GPT-5.1 leads by +12.5
FrontierMath Tier 4 (Jul 2025) · the most challenging tier of frontier mathematics, containing problems that push the absolute limits of AI mathematical reasoning.
GPT-5.1
12.5
Qwen3 235B A22B Thinking 2507
0.0
GPQA diamond
GPT-5.1 leads by +13.6
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
GPT-5.1
87.0
Qwen3 235B A22B Thinking 2507
73.4
Lmca
GPT-5.1 leads by +17.2
GPT-5.1
51.6
Qwen3 235B A22B Thinking 2507
34.5
Mystery Game Puzzles
GPT-5.1 leads by +10.8
GPT-5.1
10.8
Qwen3 235B A22B Thinking 2507
0.0
OTIS Mock AIME 2024-2025
GPT-5.1 leads by +1.9
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
GPT-5.1
88.6
Qwen3 235B A22B Thinking 2507
86.7
SimpleQA Verified
GPT-5.1 leads by +7.6
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
GPT-5.1
48.0
Qwen3 235B A22B Thinking 2507
40.4
WeirdML
GPT-5.1 leads by +19.7
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
GPT-5.1
60.8
Qwen3 235B A22B Thinking 2507
41.0
Full benchmark table
| Benchmark | GPT-5.1 | Qwen3 235B A22B Thinking 2507 |
|---|---|---|
Chatbot Arena Elo · Overall | 1438.5 | 1400.0 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 39.0 | 7.4 |
Dtbench | 83.5 | 67.1 |
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. | 31.0 | 14.9 |
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. | 12.5 | 0.0 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 87.0 | 73.4 |
Lmca | 51.6 | 34.5 |
Mystery Game Puzzles | 10.8 | 0.0 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 88.6 | 86.7 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 48.0 | 40.4 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 60.8 | 41.0 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $1.25 | $10.00 | 400K tokens (~200 books) | $34.38 | |
| $0.45 | $3.50 | 128K tokens (~64 books) | $12.13 |