!uv pip install -U transformers accelerate huggingface_hub torch torchvision torchaudio
import os
import torch
from google.colab import userdata
from huggingface_hub import HfApi
from transformers import pipeline
os.environ["HF_TOKEN"] = userdata.get("HF_TOKEN")
hf_api = HfApi()
pipe = pipeline(
"text-generation",
model="google/gemma-4-E2B-it", # 2B effective (mobile)
model="google/gemma-4-E4B-it", # 4B effective (mobile and laptops)
model="google/gemma-4-12B-it", # 12B (laptops and desktops)
model="google/gemma-4-26B-A4B-it", # 26B 4B active (desktops and servers)
model="google/gemma-4-31B-it", # 31B (servers)
dtype=torch.bfloat16,
device_map="auto"
)
messages = [
{"role": "user", "content": "Explain quantum computing simply:"}
]
output = pipe(messages, max_new_tokens=150)
print(output[0]["generated_text"][-1]["content"])