MODEL HISTORY
Claude 3.5 Haiku 指标历史
Claude 3.5 Haiku 的公开评测事实历史，共 30 条观测、26 项指标；只记录来源变化，不插值、不合成本站总分。
怎样理解这份历史
每条记录对应来源快照中的一个原始指标。相同值重复抓取不会制造新记录；只有数值、临时状态或移除状态发生变化时才追加。首个快照是本站的可追溯起点，不代表该指标第一次对外发布。
Index、Elo、百分比等单位不能互相相减，也不会在这里合成为“综合总分”。
APPEND-ONLY FACTS
观测记录
按观测时间倒序；同日多项指标分别保留。
| 观测日期 | 基准 / 版本 | 指标 | 原始值 | 状态 |
|---|---|---|---|---|
| Artificial Analysis model aggregatevpublic-model-v1 | 缓存读取价格 / 1M Tokenusd_per_1m_tokens | — | 来源移除 | |
| Artificial Analysis model aggregatevpublic-model-v1 | 缓存写入价格 / 1M Tokenusd_per_1m_tokens | — | 来源移除 | |
| Artificial Analysis model aggregatevpublic-model-v1 | 输入价格 / 1M Tokenusd_per_1m_tokens | — | 来源移除 | |
| Artificial Analysis model aggregatevpublic-model-v1 | 输出价格 / 1M Tokenusd_per_1m_tokens | — | 来源移除 | |
| Artificial Analysis capability indicesvpublic-model-v1 | Agentic Indexindex | 1.9 | 有效 | |
| Artificial Analysis model aggregatevpublic-model-v1 | 缓存读取价格 / 1M Tokenusd_per_1m_tokens | US$0.08 | 有效 | |
| Artificial Analysis model aggregatevpublic-model-v1 | 缓存写入价格 / 1M Tokenusd_per_1m_tokens | US$1 | 有效 | |
| Artificial Analysis agentic capability indexvunversioned | agentic capability indexindex | 1.9 | 有效 | |
| Artificial Analysis coding capability indexvunversioned | coding capability indexindex | 15.9 | 有效 | |
| Artificial Analysis finance capability indexvunversioned | finance capability indexindex | 9.3 | 有效 | |
| Artificial Analysis strategy capability indexvunversioned | strategy capability indexindex | 8.7 | 有效 | |
| Artificial Analysis legal capability indexvunversioned | legal capability indexindex | 13.4 | 有效 | |
| Artificial Analysis healthcare capability indexvunversioned | healthcare capability indexindex | 10.7 | 有效 | |
| Artificial Analysis engineering capability indexvunversioned | engineering capability indexindex | 17.7 | 有效 | |
| Artificial Analysis economics capability indexvunversioned | economics capability indexindex | 18.8 | 有效 | |
| Artificial Analysis capability indicesvpublic-model-v1 | Coding Indexindex | 15.9 | 有效 | |
| Artificial Analysis model aggregatevpublic-model-v1 | 上下文窗口tokens | 200,000 | 有效 | |
| Artificial Analysis intelligence component evaluationsv4.1 | CritPtfraction | 0.0% | 有效 | |
| Artificial Analysis intelligence component evaluationsv4.1 | GDPval-AAelo | 444 | 有效 | |
| Artificial Analysis intelligence component evaluationsv4.1 | GDPval-AA normalizedfraction | 0.0% | 有效 | |
| Artificial Analysis intelligence component evaluationsv4.1 | GPQA Diamondfraction | 40.8% | 有效 | |
| Artificial Analysis intelligence component evaluationsv4.1 | Humanity's Last Examfraction | 3.5% | 有效 | |
| Artificial Analysis model aggregatevpublic-model-v1 | 输入价格 / 1M Tokenusd_per_1m_tokens | US$0.8 | 有效 | |
| Artificial Analysis capability indicesvpublic-model-v1 | Intelligence Indexindex | 12.3 | 有效 | |
| Artificial Analysis intelligence component evaluationsv4.1 | AA-LCRfraction | 23.3% | 有效 | |
| Artificial Analysis intelligence component evaluationsv4.1 | AA-Omniscienceindex | -23.2 | 有效 | |
| Artificial Analysis model aggregatevpublic-model-v1 | 输出价格 / 1M Tokenusd_per_1m_tokens | US$4 | 有效 | |
| Artificial Analysis intelligence component evaluationsv4.1 | SciCodefraction | 27.4% | 有效 | |
| Artificial Analysis intelligence component evaluationsv4.1 | Tau Bankingfraction | 5.8% | 有效 | |
| Artificial Analysis intelligence component evaluationsv4.1 | Terminal-Bench v2.1fraction | 10.1% | 有效 |