Unverified50% confidenceBenchmarkExact time
在 SWE-bench 基准测试中，顶尖编码智能体解决真实 GitHub issue 平均需要 15-50 次工具调用，执行时间从几分钟到几十分钟不等
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
8/29/2026
First Seen
Valid until: 11/27/2026
Sources
Related Entities
Related Claims
UnverifiedThe SWE-bench benchmark from Princeton University in 2023 showed that the most advanced AI programming tools at the time achieved a success rate of only about 13% on real GitHub issue fix tasks.74% similarUnverifiedThe GitHub user manual project designed a learning path ranging from a 15-minute introduction to a 13-hour mastery track.72% similarUnverifiedSWE-bench要求模型在真实GitHub仓库中定位问题并生成可通过CI测试的补丁，以Resolve Rate作为核心指标71% similarUnverifiedThe GitHub open-source project OpenHuman has surpassed 9K total stars, can connect to 118 services including Gmail, GitHub, Slack, and Notion, and automatically fetches new data every 20 minutes.70% similarUnverifiedIn the GitHub accessibility agent tests, 3,535 PRs have been reviewed, with a problem resolution rate of 68%.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/816973API
curl https://kongchang.com/api/v1/knowledge/claims/816973MCP
get_claim(id=816973)