Always Prompts Insufficient Video Memory (whole all Unable to allocate)

The video memory is full, use a smaller model, such as medium small instead of large-v3. The minimum video memory requirement for large-v3 is 8G, but this does not mean that an 8G graphics card can run smoothly. Other software will also occupy video memory at the same time, and the larger the video, the more video memory is required. When this error is prompted, please try:

1. Use a smaller model, such as small/medium/base
2. If you still want to use a large model, choose to use "Pre-segmentation" or "Equal Segmentation"
3. Modify Menu Bar--Tools/Options--Advanced Options

CUDA data type=float32 change to int8, if an error occurs, change to int16
beam_size=5 change 5 to 1
best_of=5 change 5 to 1
Context change true to false