Qwen3.8-Max on Q-Bench Video
69.0%Rank #3 of 9 on Q-Bench Video
- Coverage
- 99.77% (866/868)
- Cost per task
- $0.0173
- Median latency
- 32.9s
- Tokens per task
- 1,603 in · 2,351 out
Run setting: 1 fps native video.
Tokens and latency
Tokens per call
| Token type | Mean | Median | 90th pct. |
|---|---|---|---|
| Input | 1,603 | 1,470 | 3,089 |
| of which cached | 12 | 0 | 0 |
| of which frames or video | 1,520 | 1,369 | 3,010 |
| Output | 2,351 | 1,207 | 6,502 |
Over 867 calls that returned an answer, including questions asked again after a failed attempt.
Latency
- Median
- 32.9s
- Mean
- 60.3s
- 90th percentile
- 163.7s
- 99th percentile
- 258.0s
Per call, end to end including file upload and processing.