DeepSeek R1 Distill Qwen 14B
Zero-eval
by DeepSeek
+
+
+
+
About
DeepSeek-R1-Distill-Qwen-14B was developed as a mid-sized distilled variant based on Qwen, designed to balance reasoning capability with practical deployment considerations. Built to provide strong analytical performance while remaining accessible, it serves applications requiring reliable reasoning without flagship-scale resources.
+
+
+
+
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
71.5%
Best Score
93.9%
High Performers (80%+)
2+
+
+
+
All Benchmark Results for DeepSeek R1 Distill Qwen 14B
Complete list of benchmark scores with detailed information
| MATH-500 | text | 0.94 | 93.9% | Self-reported | |
| AIME 2024 | text | 0.80 | 80.0% | Self-reported | |
| GPQA | text | 0.59 | 59.1% | Self-reported | |
| LiveCodeBench | text | 0.53 | 53.1% | Self-reported |
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+