Skip to main content
The LLaVA class provides a wrapper for the LLaVA model, which answers questions about visual media (images) using the LLaVA framework.
​
class LLaVA()
​
use_local
boolean
default:"False"
If True, inference call is run on the local VM, else offloaded onto GRID-Cortex. Defaults to False.
​
def run()
​
rgbimage
np.ndarray
required
The input RGB image of shape (M,N,3)(M,N,3).
​
prompt
str
required
The question to answer about the media.
​
Returns
str
The response to the prompt.
from grid.model.perception.vlm.llava import LLaVA
car = AirGenCar()

# We will be capturing an image from the AirGen simulator 
# and run model inference on it.

img =  car.getImage("front_center", "rgb").data

model = LLaVA(use_local = False)
result = model.run(rgbimage=img, prompt=<prompt>)
print(result)
  • License
  • Source
This code is licensed under the Apache 2.0 License.