LLM Decoding Throughput on MiniCPM 8B 64K context length 4.1
1,159.1Throughput (tok/s)
SparDA
Updated 1mo ago
Evaluation Results
| Method | Links | |
|---|---|---|
2026.06 | 1,159.1 | |
2026.06 | 1,049.9 | |
2026.06 | 1,041.9 | |
2026.06 | 1,026.5 | |
2026.06 | 937.2 | |
2026.06 | 811.8 | |
2026.06 | 756.1 | |
2026.06 | 549.4 | |
2026.06 | 470.3 | |
2026.06 | 466.2 | |
2026.06 | 350.8 | |
2026.06 | 246.5 | |
2026.06 | 234.9 | |
2026.06 | 218.3 | |
2026.06 | 213.1 | |
2026.06 | 174.1 | |
2026.06 | 166.5 | |
2026.06 | 129.1 | |
2026.06 | 94 | |
2026.06 | 63.3 | |
2026.06 | 55.2 |