Qwen3-Coder-480B-A35B-Instruct · SWE-bench Verified(Verified subset (500 problems)) · 主文证据(ev-qwen-3-coder-480b-a35b-instruct-qual-swe-verified)
Qwen3-Coder-480B-A35B-Instruct 的 SWE-bench Verified · Verified subset (500 problems) 官方出处。见《Qwen3-Coder: Agentic Coding in the World》Qwen3-Coder official blog / Scaling Long-Horizon RL / concluding paragraph。证据 ev-qwen-3-coder-480b-a35b-instruct-qual-swe-verified。
原文与定位
As a result, Qwen3-Coder achieves state-of-the-art performance among open-source models on SWE-Bench Verified without test-time scaling.
- 官方材料
- Qwen3-Coder: Agentic Coding in the World · 定位到章节
- 定位
- 主文 · Qwen3-Coder official blog / Scaling Long-Horizon RL / concluding paragraph
- 材料版本
- Official Qwen blog HTML snapshot; article datePublished confirms 2025-07-22.
- 版本公布
- 披露时间
- · 已有首发依据
- 采集时间
- 历史依据
- 暂无可靠历史版本依据
引用文本
Qwen3-Coder-480B-A35B-Instruct · SWE-bench Verified(Verified subset (500 problems)),披露记录。官方出处 Qwen3-Coder: Agentic Coding in the World(主文,Official Qwen blog HTML snapshot; article datePublished confirms 2025-07-22.)Qwen3-Coder official blog / Scaling Long-Horizon RL / concluding paragraph,https://qwenlm.github.io/blog/qwen3-coder/#scaling-long-horizon-rl。Benchlog https://benchlog.lekshome.cc/evidence/qwen-3-coder-480b-a35b-instruct/ev-qwen-3-coder-480b-a35b-instruct-qual-swe-verified/
关联成绩与条件
此摘录未关联到已收录的成绩记录。