Molmo 2 8B on Perception Test
67.4%Rank #8 of 9 on Perception Test
- Coverage
- Full (454/454)
- Cost per task
- —
- Median latency
- 0.6s
- Tokens per task
- 2,215 in · 2 out
Run setting: 1 fps.
Tokens and latency
Tokens per call
| Token type | Mean | Median | 90th pct. |
|---|---|---|---|
| Input | 2,215 | 2,234 | 3,168 |
| of which cached | 1,286 | 1,248 | 3,120 |
| Output | 2 | 2 | 2 |
Over 454 calls that returned an answer, including questions asked again after a failed attempt.
Latency
- Median
- 0.6s
- Mean
- 0.7s
- 90th percentile
- 1.0s
- 99th percentile
- 1.1s
We only consider the first question against a video for self hosted models, ignoring prefix caching latency.