Vertex AI Model Garden Complete Guide: Claude, Llama, Gemini & Imagen on GCP

Vertex AI Model Garden is Google Cloud's catalog of pre-trained models. You can use Google's own models (Gemini / Imagen / Chirp), third-party models (Anthropic Claude), and open-source models (Llama / Mistral / Gemma / DeepSeek and more, 100+ total) through a unified interface.

Model Categories

Anthropic Claude on Vertex AI

from anthropic import AnthropicVertex

client = AnthropicVertex(region="us-east5", project_id="my-project")

message = client.messages.create(
    model="claude-opus-4@20251020",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Name 3 benefits of GCP"}],
)
print(message.content[0].text)

Open Model 1-Click Deploy

# Console: Model Garden → Llama 3.3 70B → Deploy
# Or via API
from google.cloud import aiplatform as ai

model = ai.Model.upload(
    display_name="llama-3-70b",
    artifact_uri="gs://vertex-model-garden-public/llama-3-70b",
    serving_container_image_uri="us-docker.pkg.dev/vertex-ai/prediction/text-generation-inference-cu125.0-1.ubuntu2204.py310",
)

endpoint = model.deploy(
    machine_type="g2-standard-12",
    accelerator_type="NVIDIA_L4",
    accelerator_count=1,
    min_replica_count=1,
)

prediction = endpoint.predict(instances=[{"prompt": "Hello", "max_tokens": 100}])

Fine-tuning Options

Pricing Comparison (2026)

Imagen 3 (Image Generation)

from vertexai.preview.vision_models import ImageGenerationModel

model = ImageGenerationModel.from_pretrained("imagen-3.0-generate-002")
images = model.generate_images(
    prompt="A futuristic city at sunset, photorealistic",
    number_of_images=4,
    aspect_ratio="16:9",
)
images[0].save("output.png")

Comparison with Other Clouds

Google Cloud is a trademark of Google LLC, and Claude is a trademark of Anthropic, PBC. For the latest information, see the official Model Garden page.