1
LLM DeepSWE Pareto Frontier
多模型在DeepSWE基准上的帕累托前沿，揭示代码生成效率与成本平衡。
Article URL: https://deepswe.datacurve.ai/ Comments URL: https://news.ycombinator.com/item?id=49200656 Points: 1 # Comments: 0
2
Agent runtime reduces LLM turns by 80% with a higher success rate in DeepSWE
AI Agent新方案Tura，将大模型调用次数锐减83%，成功率反升16.7个百分点，极限压缩token预算。
Article URL: https://github.com/Tura-AI/tura Comments URL: https://news.ycombinator.com/item?id=48922039 Points: 2 # Comments: 1
3
DeepSWE: Measuring Frontier Coding Agents on Original, Long-Horizon Engineering Tasks
新基准DeepSWE用113个原创、长周期软件工程任务评估编码代理，避免SWE-bench的数据泄露问题。
arXiv:2607.07946v1 Announce Type: cross Abstract: DeepSWE is a benchmark of 113 original, long-horizon software engineering tasks for evaluating codin…