High Throughput Inference Recommendation: Jamba 1.5 Large
High Throughput Inference → Jamba 1.5 Large
Ikhtisar Kasus Penggunaan
Jamba 1.5 Large by Other is a recommended option for the High Throughput Inference use case.
Model Rekomendasi
| Peringkat | Vendor | Jendela Konteks | Score |
|---|
| #3 | Other | 256K | 36/100 |
Analisis Kinerja
Kinerja Benchmark
| Benchmark | Skor |
|---|
| MMLU | 84.3 |
| HUMANEVAL | 73.7 |
| GSM8K | 83.5 |
| MATH | 60.0 |
| BBH | 82.8 |
| GPQA | 50.8 |
Kelebihan
- MMLU score 84.3, strong knowledge reasoning.
- 上下文窗口 256K，支持长文本。
- 采用 MoE 混合专家架构。
Persyaratan
- 需要 other 的 API 密钥
- 输入长度须在 256K 上下文窗口内