DeepSeek: R1 0528

DeepSeek: R1 0528

deepseek · Released May 28, 2025 Efficient
38.8
Our Score

Performance Profile

Intelligence3.5Technical1.6Value7.3Content3.5
Intelligence 3.5/10
Technical 1.6/10
Content 3.5/10
Value 7.3/10

May 28th update to the original DeepSeek R1 Performance on par with OpenAI o1, but open-sourced and with fully open reasoning tokens. It's 671B parameters in size, with 37B active..

$0.50 / 1M
Input Price
$2.15 / 1M
Output Price
163,840 tokens
Context Window

Capabilities

Tool Use Function Calling

Architecture

ModalityText → Text
TokenizerDeepSeek
Instruct Typedeepseek-r1

Performance Indices

Source: Artificial Analysis

16.4 Intelligence Index
7.8 Coding Index
1.5 Agentic Index
63.7 Math Index

Benchmark Scores

Intelligence

GPQA Diamond 61.2% Graduate-level scientific reasoning
HLE 5.6% Humanity's Last Exam
MMLU Pro 73.9% Multi-task language understanding
MATH 500 93.2% Mathematical problem-solving
AIME 65% Competition mathematics
AIME 2025 63.7% Competition mathematics (2025)
SciCode 20.4% Scientific computing

Technical

LiveCodeBench 51.3% Live coding evaluation
TerminalBench Hard 1.5% Agentic terminal tasks

Content

IFBench 19.9% Instruction following
LCR 13% Long-context reasoning

Benchmark data from Artificial Analysis and Hugging Face

How does DeepSeek: R1 0528 stack up?

Compare side-by-side with other efficient models.

Compare Models

Model Information

OpenRouter ID deepseek/deepseek-r1-0528
Providerdeepseek
Model FamilyDeepSeek
Release Date May 28, 2025
Context Length163,840 tokens
Status Active

Pricing

Token Type Cost per 1M tokens Cost per 1K tokens
Input $0.50 $0.000500
Output $2.15 $0.002150

Live Performance

Live endpoint metrics — refreshed every 30 minutes.

99.9%
Avg Uptime
565ms
Best Latency (TTFT)
30 tok/s
Best Throughput
3/5
Active Endpoints
Available via: DeepInfra, SiliconFlow, AtlasCloud, Novita, Together