Saltar al contenido principal

DeepSeek R1 0528 (May '25)

DeepSeekDeepSeekOpen WeightMIT · Uso Comercial

Descripción

DeepSeek-R1-0528 is the May 28, 2025 version of DeepSeek's reasoning model. It features advanced thinking capabilities and serves as a benchmark comparison for newer models like DeepSeek-V3.1. This model excels in complex reasoning tasks, mathematical problem-solving, and code generation through its thinking mode approach.

Fecha de lanzamiento
2025-05-28
Parámetros
671.0B
Longitud del contexto
164K
Modalidades
text

Radar de capacidades

39
general
69
coding
83
reasoning
54
science
10
agents
0
multimodal

Rankings

Dominio#PosiciónPuntuaciónFuente
Capacidad agéntica146
24.0
LS
Ranking de codificación176
59.0
AA
Ranking general275
46.0
AA
Ciencia172
59.0
AA

Puntuaciones de benchmarks (LLM Stats)

(LLM Stats (zeroeval))

Agents

t2-bench80.2%Aut.
Terminal-Bench 2.0Stanford × Laude Institute (2026)46.4%Aut.
Toolathlon35.2%Aut.
BrowseCompOpenAI (2025)8.9%Aut.
Terminal-Bench5.7%Aut.

Biology

GPQANYU + Cohere + Anthropic (2023)81.0%Aut.

Code

LiveCodeBench73.3%Aut.
Aider-Polyglot71.6%Aut.
SWE-Bench Verified44.6%Aut.
SWE-bench Multilingual30.5%Aut.

Factuality

SimpleQA92.3%Aut.

Finance

MMLU-Pro85.0%Aut.

General

MMLU-Redux93.4%Aut.

Math

AIME 202491.4%Aut.
AIME 202587.5%Aut.
HMMT 202579.4%Aut.
CodeForces0.64 / 3000Aut.
Humanity's Last Exam17.7%Aut.

Reasoning

BrowseComp-zh35.7%Aut.

Índices de evaluación AA

(Artificial Analysis)
Math Index(Artificial Analysis)
76.0
Intelligence Index(Artificial Analysis)
20.4
Math 500(OpenAI (2024), subset of Hendrycks et al. MATH (2021))
1.0
Aime(MAA (Mathematical Association of America))
0.9
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))
0.8
Gpqa(NYU + Cohere + Anthropic (2023))
0.8
Livecodebench(UC Berkeley + MIT + Cornell (2024))
0.8
Aime 25(MAA (Mathematical Association of America))
0.8
Lcr(Artificial Analysis)
0.6
Scicode(UIUC + Argonne National Lab (2024))
0.4
Ifbench(Google Research (2023))
0.4
Tau2(Sierra + U Toronto + Vector Institute (2025))
0.4
Terminalbench Hard(Stanford × Laude Institute (2026))
0.2
Hle(Center for AI Safety + Scale AI (2025))
0.2

Puntuaciones por categoría LLM Stats

(LLM Stats (zeroeval))
Language
90
Factuality
90
Legal
80
Physics
80
Finance
80
Healthcare
80
Biology
80
Chemistry
80
Math
70
Reasoning
60
General
60
Code
50
Frontend Development
40
Search
20
Vision
20
Agents
10

Precios

Precio de entrada$1.35 / 1M tokens
Precio de salida$4.2 / 1M tokens
Precio mixto (3:1)$2.063 / 1M tokens
Precio de lectura caché$0.35 / 1M tokens

Velocidad

Tokens/seg0.0
Retraso del primer token0.00s
Tiempo hasta la respuesta0.00s

Ranking de Precios por Proveedor

Ranking de Precios por Proveedor

10 proveedores

Más barato: NanoGPTMás caro: Azure
ProveedorEntradaSalida
1NanoGPTMás barato
$0.4
$1.7
2OpenRouter
$0.5
$2.15
3Alibaba (China)
$0.574
$2.294
4TensorX
$0.66
$2.6
5Jiekou.AI
$0.7
$2.5
6NovitaAI
$0.7
$2.5
7Kilo Gateway
$0.7
$2.5
8DeepSeekPRINCIPAL
$1.35
$4.2
9Azure Cognitive Services
$1.35
$5.4
10Azure
$1.35
$5.4

Comparar precios entre diferentes proveedores de API para este modelo.

Fuentes externas