Compare · ModelsLive · 2 picked · head to head
DeepSeek V3 vs GPT-5.2
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
GPT-5.2 wins on 9/9 benchmarks
GPT-5.2 wins 9 of 9 shared benchmarks. Leads in arena · general · math.
Category leads
arena·GPT-5.2general·GPT-5.2math·GPT-5.2knowledge·GPT-5.2reasoning·GPT-5.2coding·GPT-5.2
Hype vs Reality
Attention vs performance
DeepSeek V3
#79 by perf·no signal
GPT-5.2
#162 by perf·#9 by attention
Best value
DeepSeek V3
15.9x better value than GPT-5.2
DeepSeek V3
92.9 pts/$
$0.60/M
GPT-5.2
5.9 pts/$
$7.88/M
Vendor risk
Mixed exposure
One or more vendors flagged
DeepSeek
$3.4B·Tier 1
OpenAI
$840.0B·Tier 1
Head to head
9 benchmarks · 2 models
DeepSeek V3GPT-5.2
Chatbot Arena Elo · Overall
GPT-5.2 leads by +77.1
DeepSeek V3
1358.4
GPT-5.2
1435.6
Dtbench
GPT-5.2 leads by +43.5
DeepSeek V3
41.3
GPT-5.2
84.9
FrontierMath-2025-02-28-Private
GPT-5.2 leads by +37.7
FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning.
DeepSeek V3
3.0
GPT-5.2
40.7
GPQA diamond
GPT-5.2 leads by +48.4
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
DeepSeek V3
43.8
GPT-5.2
92.2
Lmca
GPT-5.2 leads by +33.5
DeepSeek V3
18.2
GPT-5.2
51.7
Metr Time Horizons
GPT-5.2 leads by +27.9
DeepSeek V3
47.4
GPT-5.2
75.3
OTIS Mock AIME 2024-2025
GPT-5.2 leads by +80.4
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
DeepSeek V3
15.8
GPT-5.2
96.1
SimpleBench
GPT-5.2 leads by +32.3
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
DeepSeek V3
2.7
GPT-5.2
35.0
WeirdML
GPT-5.2 leads by +36.1
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
DeepSeek V3
36.1
GPT-5.2
72.2
Full benchmark table
| Benchmark | DeepSeek V3 | GPT-5.2 |
|---|---|---|
Chatbot Arena Elo · Overall | 1358.4 | 1435.6 |
Dtbench | 41.3 | 84.9 |
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. | 3.0 | 40.7 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 43.8 | 92.2 |
Lmca | 18.2 | 51.7 |
Metr Time Horizons | 47.4 | 75.3 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 15.8 | 96.1 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 2.7 | 35.0 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 36.1 | 72.2 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $0.32 | $0.89 | 164K tokens (~82 books) | $4.63 | |
| $1.75 | $14.00 | 400K tokens (~200 books) | $48.13 |