Qwen3-Coder
Alibaba·
LLMsopen-weightEdge / Consumer
Overview
Code-specialized open model optimized for software engineering. Achieved top-tier SWE-bench scores among open-weight models.
Capabilities and innovations
Code-specialized architectureTop SWE-bench open-weight performanceMulti-language code generationRepository-level understandingCode-optimized training pipelineRepository-aware context handlingSpecialized coding reward model
Benchmarks
| Benchmark | Score | Source |
|---|---|---|
| SWE-bench Verified | 52.7% | unsourced |
Architecture and hardware
- Parameters
- 32B total (Dense)
- Estimated VRAM at Q4
- ~19 GB, Edge / Consumer class
- Quantization formats
- GGUF, GPTQ, AWQ
- Recommended runtime
- Ollama
- License
- Apache 2.0
Links
More from Alibaba
Data curated by AI Model Timeline. See the methodology for admission criteria, benchmark eras and source priority.