Compare · ModelsLive · 2 picked · head to head
GPT-5.1 vs DeepSeek V3.2
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 reasoning · arena · knowledge.
Category leads
reasoning·GPT-5.1arena·GPT-5.1knowledge·GPT-5.1math·GPT-5.1coding·GPT-5.1
Hype vs Reality
Attention vs performance
GPT-5.1
#97 by perf·no signal
DeepSeek V3.2
#84 by perf·no signal
Best value
DeepSeek V3.2
19.1x better value than GPT-5.1
GPT-5.1
8.8 pts/$
$5.63/M
DeepSeek V3.2
168.3 pts/$
$0.32/M
Vendor risk
Mixed exposure
One or more vendors flagged
OpenAI
$840.0B·Tier 1
DeepSeek
$3.4B·Tier 1
Head to head
11 benchmarks · 2 models
GPT-5.1DeepSeek V3.2
ARC-AGI
GPT-5.1 leads by +15.8
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
GPT-5.1
72.8
DeepSeek V3.2
57.0
ARC-AGI-2
GPT-5.1 leads by +13.6
ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data.
GPT-5.1
17.6
DeepSeek V3.2
4.0
Chatbot Arena Elo · Coding
GPT-5.1 leads by +11.9
GPT-5.1
1338.8
DeepSeek V3.2
1326.9
Chatbot Arena Elo · Overall
GPT-5.1 leads by +14.1
GPT-5.1
1438.5
DeepSeek V3.2
1424.4
Chess Puzzles
GPT-5.1 leads by +18.0
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
GPT-5.1
32.0
DeepSeek V3.2
14.0
FrontierMath-2025-02-28-Private
GPT-5.1 leads by +8.9
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
DeepSeek V3.2
22.1
FrontierMath-Tier-4-2025-07-01-Private
GPT-5.1 leads by +10.4
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
DeepSeek V3.2
2.1
GPQA diamond
GPT-5.1 leads by +5.6
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
83.5
DeepSeek V3.2
77.9
OTIS Mock AIME 2024-2025
GPT-5.1 leads by +0.8
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
GPT-5.1
88.6
DeepSeek V3.2
87.8
SimpleQA Verified
GPT-5.1 leads by +21.4
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
GPT-5.1
48.9
DeepSeek V3.2
27.5
Terminal Bench
GPT-5.1 leads by +8.0
Terminal-Bench 2.0 · evaluates AI agents on real terminal-based coding tasks · writing scripts, debugging, running tests, and managing projects entirely through command-line interaction. Tests both code quality and terminal fluency. Claude Opus 4.7 scores 69.4%, demonstrating significant agentic terminal competence.
GPT-5.1
47.6
DeepSeek V3.2
39.6
Full benchmark table
| Benchmark | GPT-5.1 | DeepSeek V3.2 |
|---|---|---|
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 72.8 | 57.0 |
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. | 17.6 | 4.0 |
Chatbot Arena Elo · Coding | 1338.8 | 1326.9 |
Chatbot Arena Elo · Overall | 1438.5 | 1424.4 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 32.0 | 14.0 |
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 | 22.1 |
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 | 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. | 83.5 | 77.9 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 88.6 | 87.8 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 48.9 | 27.5 |
Terminal Bench Terminal-Bench 2.0 · evaluates AI agents on real terminal-based coding tasks · writing scripts, debugging, running tests, and managing projects entirely through command-line interaction. Tests both code quality and terminal fluency. Claude Opus 4.7 scores 69.4%, demonstrating significant agentic terminal competence. | 47.6 | 39.6 |
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 | |
| $0.25 | $0.38 | 131K tokens (~66 books) | $2.83 |