Qwen2.5-Coder 7B Instruct
by Alibaba / Qwen
+
+
+
+
About
Qwen2.5-Coder-7B-Instruct is a 7-billion-parameter code-specialized model from Alibaba, released in November 2024 as part of the Qwen2.5-Coder family, trained on a curated corpus spanning 92 programming languages with emphasis on code generation, debugging, and fill-in-the-middle completion. Built on the Qwen2.5 architecture, it extends the base series' improvements in instruction-following and long-context handling to coding-specific tasks within a compact deployable footprint. It became popular for integration into IDE extensions, CI pipelines, and self-hosted code assistant tools.
+
+
+
+
Timeline
ReleasedNov 12, 2024
+
+
+
+
License & Family
License
Apache 2.0
Performance Overview
Performance metrics and category breakdown
Overall Performance
1 benchmarks
Average Score
83.5%
Best Score
83.5%
High Performers (80%+)
1Top Categories
Coding
83.5%
+
+
+
+
All Benchmark Results for Qwen2.5-Coder 7B Instruct
Complete list of benchmark scores with detailed information
| MBPP | Coding | 83.50 | 83.5% | Unverified |