twelvelabs-pegasus1.5

A multimodal model that provides comprehensive video understanding and analysis capabilities, including video-to-text generation, structured video segmentation, image-guided prompting, and timestamp-level temporal grounding.

Details

• Model name: twelvelabs-pegasus1.5

• Model source: https://www.twelvelabs.io

• Model author: Twelve Labs, Inc.

• Model license: https://www.twelvelabs.io/legal/tos

• Exposes embeddings? no

• Tags: video, caption, official

Requirements

• Packages: twelvelabs>=1.2.8

• CPU support

  • N/A

• GPU support

  • N/A

Example usage

import fiftyone as fo
import fiftyone.zoo as foz

dataset = foz.load_zoo_dataset("quickstart-video")

# Default configuration
model = foz.load_zoo_model("twelvelabs-pegasus1.5")

# Custom prompt and generation length
model = foz.load_zoo_model(
    "twelvelabs-pegasus1.5",
    prompt="List the main objects that appear in this video.",
    max_tokens=1024,
)

dataset.apply_model(model, label_field="caption")

session = fo.launch_app(dataset)