Text Generation Throughput on 64K Serving Scenario (Single H100 GPU)
0.8Throughput (K tok/s)gpt-oss-puzzle-88B
Evaluation Results
| Method | Links | |
|---|---|---|
| gpt-oss-puzzle-88BKV precision=KV FP82026.02 | 0.8 | |
| HyperNova-60BKV precision=KV BF162026.02 | 0.7 | |
| gpt-oss-puzzle-88BKV precision=KV BF162026.02 | 0.5 | |
| gpt-oss-120BKV precision=KV FP82026.02 | 0.3 | |
| gpt-oss-120BKV precision=KV BF162026.02 | 0.2 |