{"version":"2026-10-19.observations.2","checkedAt":"2026-10-19","series":{"id":"lcb-code","group":"capability","title":"历史编程题：LiveCodeBench overall","question":"在这组历史编程题上，一次生成的执行通过成绩是什么？","unit":"题目pass@1 %","denominator":"同默认题目窗口454题；一次生成执行评分","protocol":"窗口边界2024-08-01至2025-05-01；实际题目2024-08-04至2025-04-07；只选overall；pass@1估计单次候选代码通过测试的表现，k=1不等于各模型只生成一次；逐行实际生成次数未知","scope":"Oct14数据快照，题目窗口2025截止；模型日/运行日未知","summary":"精选6个模型/档位的公布值介于27.20至80.18（题目pass@1 %）。这是历史题集横截面。不能标作2026新题表现，也不能推导真实维护项目的净收益。","nextAction":"先确认题型是否像你的任务；实际代码仍需目标行为、原行为与回归验收。","rows":[{"id":"obs-0376","label":"Claude-3.5-Sonnet-20241022","value":36.38766519823788,"rawValue":"36.38766519823788","period":null,"modelDate":null,"modelVersion":"Claude-3.5-Sonnet-20241022","n":454,"ci":null,"cost":null,"locator":"performances_generation.json；model=Claude-3.5-Sonnet-20241022；overall 默认窗口","notes":"实际题日期2024-08-04至2025-04-07；发布日无法可靠对齐，保持未知；执行评分；逐行工具/预算不完整","valueStatus":"reported","sourceId":"C-LCB","actualCoverage":null,"runAt":null,"scorePublishedAt":null,"toolBudget":null},{"id":"obs-0377","label":"Claude-Opus-4","value":46.91629955947136,"rawValue":"46.91629955947136","period":null,"modelDate":null,"modelVersion":"Claude-Opus-4","n":454,"ci":null,"cost":null,"locator":"performances_generation.json；model=Claude-Opus-4；overall 默认窗口","notes":"实际题日期2024-08-04至2025-04-07；发布日无法可靠对齐，保持未知；执行评分；逐行工具/预算不完整","valueStatus":"reported","sourceId":"C-LCB","actualCoverage":null,"runAt":null,"scorePublishedAt":null,"toolBudget":null},{"id":"obs-0378","label":"DeepSeek-V3","value":27.202643171806166,"rawValue":"27.202643171806166","period":null,"modelDate":null,"modelVersion":"DeepSeek-V3","n":454,"ci":null,"cost":null,"locator":"performances_generation.json；model=DeepSeek-V3；overall 默认窗口","notes":"实际题日期2024-08-04至2025-04-07；发布日无法可靠对齐，保持未知；执行评分；逐行工具/预算不完整","valueStatus":"reported","sourceId":"C-LCB","actualCoverage":null,"runAt":null,"scorePublishedAt":null,"toolBudget":null},{"id":"obs-0379","label":"O3 (High)","value":75.77092511013215,"rawValue":"75.77092511013215","period":null,"modelDate":null,"modelVersion":"O3 (High)","n":454,"ci":null,"cost":null,"locator":"performances_generation.json；model=O3 (High)；overall 默认窗口","notes":"实际题日期2024-08-04至2025-04-07；发布日无法可靠对齐，保持未知；执行评分；逐行工具/预算不完整","valueStatus":"reported","sourceId":"C-LCB","actualCoverage":null,"runAt":null,"scorePublishedAt":null,"toolBudget":null},{"id":"obs-0380","label":"O4-Mini (High)","value":80.1762114537445,"rawValue":"80.1762114537445","period":null,"modelDate":null,"modelVersion":"O4-Mini (High)","n":454,"ci":null,"cost":null,"locator":"performances_generation.json；model=O4-Mini (High)；overall 默认窗口","notes":"实际题日期2024-08-04至2025-04-07；发布日无法可靠对齐，保持未知；执行评分；逐行工具/预算不完整","valueStatus":"reported","sourceId":"C-LCB","actualCoverage":null,"runAt":null,"scorePublishedAt":null,"toolBudget":null},{"id":"obs-0381","label":"Qwen3-235B-A22B","value":65.85903083700441,"rawValue":"65.85903083700441","period":null,"modelDate":null,"modelVersion":"Qwen3-235B-A22B","n":454,"ci":null,"cost":null,"locator":"performances_generation.json；model=Qwen3-235B-A22B；overall 默认窗口","notes":"实际题日期2024-08-04至2025-04-07；发布日无法可靠对齐，保持未知；执行评分；逐行工具/预算不完整","valueStatus":"reported","sourceId":"C-LCB","actualCoverage":null,"runAt":null,"scorePublishedAt":null,"toolBudget":null}],"sourceIds":["C-LCB"],"xKind":"label","connect":false,"readingIds":["reading-cost"],"taskId":null,"boundaryTasks":["C03"],"actualCoverage":null,"evidenceStatus":"active"},"sources":[{"id":"C-LCB","title":"LiveCodeBench历史代码生成窗口","url":"https://livecodebench.github.io/leaderboard.html","version":"2026-10-14取得的同一公开JSON；仅overall","locator":"performances_generation.json；默认筛选边界2024-08-01至2025-05-01","scope":"454题；实际题目日期2024-08-04至2025-04-07；execution-based pass@1","license":"代码MIT；成绩JSON数据许可未单独确认；仅有限事实摘要","limit":"不是2026新题实测；模型发布日未可靠建立，全部保持null；不与AIME合并排名","publishedAt":null,"readAt":"2026-10-15","collectedAt":"2026-10-14","status":"active","nextReview":"2027-01-13","methods":["https://github.com/LiveCodeBench/LiveCodeBench"],"family":"C-LCB"}],"temporalRule":"观察期可画同来源趋势；模型发布日只展开当前横截面，未知运行/首次公开日不补。使用份额/企业采用/测试分不互换，不据此生成职业AI化比例。","actualCoverage":null}