Grok 3 Mini Beta
来自 xAI · 发布于 2025-04-09
64.8
平均分
$0.30/1M
输入价格
$0.50/1M
输出价格
131K tokens (~66 books)
上下文窗口
text
类型
Tested on 7 benchmarks with 64.8% average. Top scores: Chatbot Arena Elo — Overall (1357.4%), HELM — IFEval (95.1%), HELM — MMLU-Pro (79.9%).
基准测试分数
| 基准测试 | 类别 | 分数 | Bar |
|---|---|---|---|
| Chatbot Arena Elo — Overall | arena | 1357.4 | |
| HELM — IFEval | language | 95.1 | |
| HELM — MMLU-Pro | knowledge | 79.9 | |
| HELM — GPQA | knowledge | 67.5 | |
| HELM — WildBench | reasoning | 65.1 | |
| Aider polyglot | coding | 49.3 | |
| HELM — Omni-MATH | math | 31.8 |