Unverified50% confidenceBenchmarkExact time
2023-2024年间，部分模型在SWE-bench Verified基准上的问题解决率突破50%
1
Sources
50%
Confidence
Long-term
Relevance
7/16/2026
First Seen
Sources
Related Claims
UnverifiedThe model's rate of solving real GitHub issues on SWE-Bench rose from below 5% in early 2023 to surpassing 50% by the end of 2024.80% similarUnverifiedAs of early 2025, the pass rate of the best-performing solutions on SWE-bench is approximately 60%.74% similarUnverifiedAnalyst firms such as Gartner predict that by the end of 2025, more than 50% of enterprise-level AI inference will be completed on edge or endpoint devices.73% similarUnverifiedGartner预测到2027年超过50%的企业AI推理工作负载将在本地或边缘环境中完成72% similarUnverifiedIn early 2023, accurately analyzing a 50-page PDF document with AI was still a technical challenge; by 2024, with the maturation of long-context window models and multimodal capabilities, it had become a basic feature.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/533396API
curl https://kongchang.com/api/v1/knowledge/claims/533396MCP
get_claim(id=533396)