Gemini 2.0 Flash Thinking (Jan 2025)
来自 Google DeepMind · 发布于 2024-01-01
37.7
平均分
N/A
输入价格
N/A
输出价格
N/A
上下文窗口
text
类型
Tested on 7 benchmarks with 37.7% average. Top scores: Lech Mazur Writing (73.8%), OTIS Mock AIME 2024-2025 (57.7%), Fiction.LiveBench (52.8%).
基准测试分数
| 基准测试 | 类别 | 分数 | Bar |
|---|---|---|---|
| Lech Mazur Writing | knowledge | 73.8 | |
| OTIS Mock AIME 2024-2025 | math | 57.7 | |
| Fiction.LiveBench | knowledge | 52.8 | |
| GPQA diamond | knowledge | 42.8 | |
| Aider polyglot | coding | 18.2 | |
| SimpleBench | reasoning | 16.8 | |
| HLE | knowledge | 1.9 |