deploy_llama4.py

gistfile1.txt

# 1. Install dependencies (run in terminal)
# pip install 'google-cloud-aiplatform>=1.84.0' 'openai' 'google-auth' 'requests'

# 2. Import libraries
import vertexai
from vertexai.preview import model_garden

# 3. Initialize Vertex AI SDK (replace placeholders)
vertexai.init(project="your-project", location="your-region")

# 4. Get Llama 4 model reference from Model Garden
llama4_model = model_garden.OpenModel("meta/llama4@llama-4-scout-17b-16e-instruct")

# 5. Deploy model to an endpoint (requires EULA acceptance)
llama4_endpoint = llama4_model.deploy(accept_eula=True)

# 6. Send prediction request (prompt + parameters)
prediction = llama4_endpoint.predict(
    instances=[{
        "prompt": "Hello Llama 4! How are you?",
        "temperature": 0.7,
        "max_tokens": 50
    }]
)

# 7. Print prediction result
print(prediction.predictions[0])

# --- Example I/O ---
# Prompt:
# Hello Llama 4! How are you?
# Output:
# I'm doing well, thanks for asking!...