DeepSeek V3 0324
DeepSeekDeepSeek오픈 웨이트MIT + Model License (Commercial use allowed)
설명
A powerful Mixture-of-Experts (MoE) language model with 671B total parameters (37B activated per token). Features Multi-head Latent Attention (MLA), auxiliary-loss-free load balancing, and multi-token prediction training. Pre-trained on 14.8T tokens with strong performance in reasoning, math, and code tasks.
출시일
2025-03-25
파라미터
671.0B
컨텍스트 길이
164K
모달리티
text
능력 레이더
31
general
30
coding
54
reasoning
44
science
47
agents
0
multimodal
랭킹
벤치마크 점수 (LLM Stats)
(LLM Stats (zeroeval))Language
MMLU-Pro
81.2%자체 보고
Math
MATH-500
94.0%자체 보고
AIME 2024
59.4%자체 보고
Reasoning
GPQANYU + Cohere + Anthropic (2023)
68.4%자체 보고
LiveCodeBench
49.2%자체 보고
AA 평가 지수
(Artificial Analysis)Math 500(OpenAI (2024), subset of Hendrycks et al. MATH (2021))94.2
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))81.9
Gpqa(NYU + Cohere + Anthropic (2023))65.5
Aime(MAA (Mathematical Association of America))52.0
Tau2(Sierra + U Toronto + Vector Institute (2025))47.1
Ifbench(Google Research (2023))41.0
Math Index(Artificial Analysis)41.0
Aime 25(MAA (Mathematical Association of America))41.0
Lcr(Artificial Analysis)40.7
Livecodebench(UC Berkeley + MIT + Cornell (2024))40.5
Scicode(UIUC + Argonne National Lab (2024))39.0
Coding Index(Artificial Analysis)21.2
Terminalbench Hard(Stanford × Laude Institute (2026))15.2
Terminalbench V2 113.9
Intelligence Index(Artificial Analysis)9.7
Tau Banking4.7
Hle(Center for AI Safety + Scale AI (2025))4.7
LLM Stats 카테고리 점수
(LLM Stats (zeroeval))Language80
Legal80
Math80
Finance80
Healthcare80
Physics70
Reasoning70
General70
Biology70
Chemistry70
Code50
가격
입력 가격$0.24 / 1M 토큰
출력 가격$0.9 / 1M 토큰
혼합 가격 (3:1)$0.405 / 1M 토큰
캐시 읽기 가격$0.11 / 1M 토큰
속도
토큰/초0.0
첫 토큰 지연0.00s
첫 응답 지연0.00s
공급자 가격 순위
공급자 가격 순위
5개 공급자
최저가: DeepInfra최고가: Kilo Gateway
공급자입력출력
1DeepInfra최저가
$0
$0
2NanoGPT
$0.2
$0.77
3DeepSeek주요
$0.24
$0.9
4OpenRouter
$0.29
$1.14
5Kilo Gateway
$0.29
$1.14
이 모델의 다양한 API 공급자 간 가격 비교.