DeepSeek-R1-0528
Zero-eval
#1MMLU-Redux
#1CNMO 2024
#1MMLU-Pro
+3 more
by DeepSeek
About
DeepSeek-R1-0528 is a language model developed by DeepSeek. It achieves strong performance with an average score of 68.1% across 16 benchmarks. It excels particularly in MMLU-Redux (93.4%), AIME 2024 (91.4%), AIME 2025 (87.5%). It supports a 262K token context window for handling large documents. The model is available through 3 API providers. Released in 2025, it represents DeepSeek's latest advancement in AI technology.
Pricing Range
Input (per 1M)$0.50 -$0.70
Output (per 1M)$2.15 -$2.50
Providers3
Timeline
AnnouncedMay 28, 2025
ReleasedMay 28, 2025
Specifications
Training Tokens14.8T
License & Family
License
MIT License
Base ModelDeepSeek-V3
Benchmark Performance Overview
Performance metrics and category breakdown
Overall Performance
16 benchmarks
Average Score
68.1%
Best Score
93.4%
High Performers (80%+)
7Performance Metrics
Max Context Window
262.1KAvg Throughput
30.7 tok/sAvg Latency
1msTop Categories
code
73.3%
general
69.2%
agents
58.7%
Benchmark Performance
Top benchmark scores with normalized values (0-100%)
Ranking Across Benchmarks
Position relative to other models on each benchmark
MMLU-Redux
Rank #1 of 13
#1DeepSeek-R1-0528
93.4%
#2Qwen3-235B-A22B-Instruct-2507
93.1%
#3DeepSeek-R1
92.9%
#4Kimi K2 Instruct
92.7%
AIME 2024
Rank #6 of 41
#3o3
91.6%
#4Gemini 2.5 Pro
92.0%
#5Grok-3
93.3%
#6DeepSeek-R1-0528
91.4%
#7Gemini 2.5 Flash
88.0%
#8o3-mini
87.3%
#9DeepSeek R1 Distill Llama 70B
86.7%
AIME 2025
Rank #9 of 36
#6Gemini 2.5 Pro Preview 06-05
88.0%
#7Grok-3 Mini
90.8%
#8GPT-5 mini
91.1%
#9DeepSeek-R1-0528
87.5%
#10o3
86.4%
#11GPT-5 nano
85.2%
#12Gemini 2.5 Pro
83.0%
CNMO 2024
Rank #1 of 4
#1DeepSeek-R1-0528
86.9%
#2DeepSeek-R1
78.8%
#3Kimi K2 Instruct
74.3%
#4DeepSeek-V3
43.2%
MMLU-Pro
Rank #1 of 60
#1DeepSeek-R1-0528
85.0%
#2DeepSeek-R1
84.0%
#3Qwen3-235B-A22B-Instruct-2507
83.0%
#4DeepSeek-V3 0324
81.2%
All Benchmark Results for DeepSeek-R1-0528
Complete list of benchmark scores with detailed information
MMLU-Redux MMLU-Redux benchmark | general | text | 0.93 | 93.4% | Self-reported |
AIME 2024 AIME 2024 benchmark | general | text | 0.91 | 91.4% | Self-reported |
AIME 2025 AIME 2025 benchmark | general | text | 0.88 | 87.5% | Self-reported |
CNMO 2024 CNMO 2024 benchmark | general | text | 0.87 | 86.9% | Self-reported |
MMLU-Pro MMLU-Pro benchmark | general | text | 0.85 | 85.0% | Self-reported |
FRAMES FRAMES benchmark | general | text | 0.83 | 83.0% | Self-reported |
GPQA GPQA benchmark | general | text | 0.81 | 81.0% | Self-reported |
HMMT 2025 HMMT 2025 benchmark | general | text | 0.79 | 79.4% | Self-reported |
LiveCodeBench LiveCodeBench benchmark | code | text | 0.73 | 73.3% | Self-reported |
Aider-Polyglot Aider-Polyglot benchmark | general | text | 0.72 | 71.6% | Self-reported |
Showing 1 to 10 of 16 benchmarks