MLA Decode Latency on AMD MI300X Instance 1
1,653.8Decode Latency (µs)1st human
Evaluation Results
| Method | Links | |
|---|---|---|
| 1st humanModel=N/A2026.01 | 1,653.8 | |
| 2nd humanModel=N/A2026.01 | 1,662.8 | |
| TTT-DiscoverModel=gpt-oss-120b2026.01 | 1,669.1 | |
| 3rd humanModel=N/A2026.01 | 1,723 | |
| 4th humanModel=N/A2026.01 | 1,768.7 | |
| 5th humanModel=N/A2026.01 | 2,038.6 | |
| Best-of-25600Model=gpt-oss-120b2026.01 | 2,286 |