DeepSeek

DeepSeek-R1

by DeepSeek

+
+
+
+
About

DeepSeek-R1, released by DeepSeek on January 20, 2025, is a large reasoning model with 671 billion total parameters (37 billion active in its MoE architecture) designed for extended chain-of-thought reasoning. It features a 128K token context window and demonstrated strong performance on mathematics, coding, and scientific reasoning benchmarks at its release. DeepSeek-R1 targets complex analytical tasks, competitive programming, and applications requiring deep deliberative reasoning under an open MIT license.

+
+
+
+
Pricing Range
Input (per 1M)$0.55 -$0.55
Output (per 1M)$2.19 -$2.19
Providers1
+
+
+
+
Timeline
ReleasedJan 20, 2025
+
+
+
+
License & Family
License
MIT
Performance Overview
Performance metrics and category breakdown

Overall Performance

2 benchmarks
Average Score
32.8%
Best Score
62.4%
High Performers (80%+)
0

Performance Metrics

Max Context Window
136.0K

Top Categories

Science
62.4%
Agents
3.1%
+
+
+
+
All Benchmark Results for DeepSeek-R1
Complete list of benchmark scores with detailed information
GPQA Diamond
Science
62.40
62.4%
Unverified
BrowseComp
Agents
3.10
3.1%
Unverified