Unverified50% confidenceBenchmarkExact time
在 Terminal Bench 2.1 测试中，OpenCode 对比里 89 个任务的失败数从 29 降到 16，多解决 13 题，分数提升 14.61 个百分点
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
8/27/2026
First Seen
Valid until: 11/25/2026
Sources
Related Entities
Related Claims
UnverifiedGoogle's engineering practice report indicates that approximately 16% of test failures in large codebases are flaky failures (non-deterministic failures).66% similarUnverifiedThe task success rate of mainstream AI generation APIs is typically between 92% and 98%, meaning 2–8 out of every 100 tasks may fail.62% similarUnverified在一个396个文件的大型混合职责项目上测试时，召回率几乎跌到零62% similarUnverifiedOn the Terminal Bench benchmark, a model whose underlying weights were unchanged but whose framework was improved rose in ranking from outside the top thirty to the top five.62% similarUnverifiedIn the Agency tool-calling test, both models scored 95 points, and the only failure for both was task No. 13 (USD to JPY foreign currency conversion).62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/808665API
curl https://kongchang.com/api/v1/knowledge/claims/808665MCP
get_claim(id=808665)