Readme
The parameters: video and prompt are necessary; you can change the temperature (lower - less hallucinated but more repetitive output), but default value of 1 is fine; max_tokens you can increase if the output is cut off.
tensorsense
/
vcba_dev_v01
The model is trained to pick up nonverbal behaviors and detect deception, but can detect a lot of other things.
- Public
- 47 runs
Run time and cost
This model runs on Nvidia T4 GPU hardware. We don't yet have enough runs of this model to provide performance information.