unsloth

/

Llama-3.2-11B-Vision-Instruct-bnb-4bit

like 43

Follow

Unsloth AI 588

I've been working with a quantized model, and it has produced very satisfying results so far. However, I've noticed that the results are not always consistent, even when I set the seed.

Here is the code I'm using to set the seed:

import numpy as np
import torch
seed = 0
torch.manual_seed(seed)
set_seed(seed)

# Set seed for Python's random module
random.seed(seed)

# Set seed for NumPy
np.random.seed(seed)

# Set seed for PyTorch (if using a GPU, use the optional torch.cuda.manual_seed)
torch.manual_seed(seed)
if torch.cuda.is_available():
    torch.cuda.manual_seed(seed)```

Is it normal for the results to still vary despite setting the seed?