Compare · ModelsLive · 2 picked · head to head
R1 0528 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 reasoning · arena · knowledge.
Category leads
reasoning·GPT-5.2arena·GPT-5.2knowledge·GPT-5.2math·GPT-5.2coding·GPT-5.2
Hype vs Reality
Attention vs performance
R1 0528
#63 by perf·#16 by attention
GPT-5.2
#162 by perf·#9 by attention
Best value
R1 0528
7.5x better value than GPT-5.2
R1 0528
43.7 pts/$
$1.32/M
GPT-5.2
5.9 pts/$
$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
R1 0528GPT-5.2
ARC-AGI
GPT-5.2 leads by +65.0
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
R1 0528
21.2
GPT-5.2
86.2
ARC-AGI-2
GPT-5.2 leads by +51.8
ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data.
R1 0528
1.1
GPT-5.2
52.9
Chatbot Arena Elo · Overall
GPT-5.2 leads by +13.9
R1 0528
1421.6
GPT-5.2
1435.6
DeepResearch Bench
GPT-5.2 leads by +6.0
DeepResearch Bench · evaluates AI on complex multi-step research tasks requiring information gathering, synthesis, and producing comprehensive analyses.
R1 0528
35.1
GPT-5.2
41.1
GPQA diamond
GPT-5.2 leads by +23.8
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
R1 0528
68.4
GPT-5.2
92.2
OTIS Mock AIME 2024-2025
GPT-5.2 leads by +29.8
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
R1 0528
66.4
GPT-5.2
96.1
SimpleBench
GPT-5.2 leads by +6.0
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
R1 0528
29.0
GPT-5.2
35.0
SimpleQA Verified
GPT-5.2 leads by +11.5
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
R1 0528
27.4
GPT-5.2
38.9
WeirdML
GPT-5.2 leads by +30.6
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
R1 0528
41.6
GPT-5.2
72.2
Full benchmark table
| Benchmark | R1 0528 | GPT-5.2 |
|---|---|---|
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 21.2 | 86.2 |
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. | 1.1 | 52.9 |
Chatbot Arena Elo · Overall | 1421.6 | 1435.6 |
DeepResearch Bench DeepResearch Bench · evaluates AI on complex multi-step research tasks requiring information gathering, synthesis, and producing comprehensive analyses. | 35.1 | 41.1 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 68.4 | 92.2 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 66.4 | 96.1 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 29.0 | 35.0 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 27.4 | 38.9 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 41.6 | 72.2 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
People also compared