DeepSeek

DeepSeek R1 Distill Qwen 7B

Zero-eval

by DeepSeek

+
+
+
+
About

DeepSeek-R1-Distill-Qwen-7B was developed as a compact distilled model, designed to provide reasoning capabilities in an efficient 7B parameter package. Built to extend analytical AI to a broad range of applications, it balances capability with the practical benefits of a smaller model size.

+
+
+
+
Timeline
AnnouncedJan 20, 2025
ReleasedJan 20, 2025
+
+
+
+
Specifications
Training Tokens14.8T
+
+
+
+
License & Family
License
MIT
Performance Overview
Performance metrics and category breakdown

Overall Performance

4 benchmarks
Average Score
65.7%
Best Score
92.8%
High Performers (80%+)
2
+
+
+
+
All Benchmark Results for DeepSeek R1 Distill Qwen 7B
Complete list of benchmark scores with detailed information
MATH-500
text
0.93
92.8%
Self-reported
AIME 2024
text
0.83
83.3%
Self-reported
GPQA
text
0.49
49.1%
Self-reported
LiveCodeBench
text
0.38
37.6%
Self-reported
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+