Literature Review Recommendation: Claude 3.5 Sonnet (2024-10-22)
Literature Review → Claude 3.5 Sonnet (2024-10-22)
Ikhtisar Kasus Penggunaan
Claude 3.5 Sonnet (2024-10-22) by Anthropic is a recommended option for the Literature Review use case.
Model Rekomendasi
| Peringkat | Vendor | Jendela Konteks | Score |
|---|
| #3 | Anthropic | 200K | 37/100 |
Analisis Kinerja
Kinerja Benchmark
| Benchmark | Skor |
|---|
| MMLU | 87.2 |
| HUMANEVAL | 86.1 |
| GSM8K | 94.0 |
| MATH | 75.1 |
| BBH | 84.2 |
| GPQA | 50.9 |
Kelebihan
- MMLU score 87.2, strong knowledge reasoning.
- HumanEval 86.1, excellent code generation.
- GSM8K 94.0, robust math reasoning.
- 上下文窗口 200K，支持长文本。
Persyaratan
- 需要 anthropic 的 API 密钥
- 输入长度须在 200K 上下文窗口内