R1
오픈소스제공 DeepSeek · 출시일 2025-01-20
45.1
평균 점수
$0.70/1M
입력 가격
$2.50/1M
출력 가격
64K tokens (~32 books)
컨텍스트 윈도우
text
유형
Tested on 14 benchmarks with 45.1% average. Top scores: Chatbot Arena Elo — Overall (1397.5%), MATH level 5 (93.0%), Lech Mazur Writing (83.0%).
벤치마크 점수
| 벤치마크 | 카테고리 | 점수 | Bar |
|---|---|---|---|
| Chatbot Arena Elo — Overall | arena | 1397.5 | |
| MATH level 5 | math | 93.0 | |
| Lech Mazur Writing | knowledge | 83.0 | |
| Fiction.LiveBench | knowledge | 69.4 | |
| GPQA diamond | knowledge | 62.3 | |
| Aider polyglot | coding | 56.9 | |
| OTIS Mock AIME 2024-2025 | math | 53.3 | |
| WeirdML | coding | 36.5 | |
| DeepResearch Bench | knowledge | 35.1 | |
| Balrog | knowledge | 34.9 | |
| SimpleQA Verified | knowledge | 27.4 | |
| SimpleBench | reasoning | 17.1 | |
| ARC-AGI | reasoning | 15.8 | |
| ARC-AGI-2 | reasoning | 1.3 |