Model Garden (Gemini Enterprise Agent Platform)
Google CloudGoogle Cloud's model catalog on Gemini Enterprise Agent Platform, formerly Vertex AI, listing Google's own models, partner models such as Anthropic's Claude, and open models including Gemma, Llama, Mistral and popular Hugging Face repositories. Each entry can be tested, tuned and deployed to a managed endpoint in one click or through the SDK, CLI or REST API, and customers can deploy their own weights into their project.
Kept under its Vertex AI slug. Google's pages now call the platform Gemini Enterprise Agent Platform and the catalog Model Garden, and the old Vertex AI documentation URLs redirect there.
Openness
1 high confidence- license
- proprietary(hosted service
- source
- closed(managed Google Cloud service)
Model Garden is part of Google Cloud's managed AI platform and has no self-hostable form. Many of the models it lists are open, but the catalog and the deployment service around them are Google's.
- https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-garden/explore-models recorded 2026-09-27
"Serving generative AI models can be difficult—Gemini Enterprise Agent Platform handles model deployment and serving for you"; open models are billed for "the compute resources used to deploy the model to an endpoint."
Adoption
not assessedGoogle publishes the size of the catalog, not how many customers use it or how often its models are deployed. A catalog count measures what is offered, so no reading was recorded.
- https://cloud.google.com/model-garden recorded 2026-09-27
"Our curated set of 200+ available models are best in class for their category"; no usage figure on the page.
Capability
3 high confidenceModel Garden gathers models from Google, partners and the open ecosystem into one catalog and deploys any of them to a managed endpoint. Google decides what is listed, and a customer's own weights stay in that customer's project rather than being published for others, so it is a curated catalog rather than an open registry like Hugging Face Hub.
- https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/model-garden/deploy-models-with-custom-weights recorded 2026-09-27
"You can deploy your custom models using the custom weights import by uploading your model artifacts to a Cloud Storage bucket in your project"
- https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/open-models/deploy-model-garden recorded 2026-09-27
"you can deploy the model to a Gemini Enterprise Agent Platform Endpoint by using one-click deployment."
- https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/open-models/use-hugging-face-models recorded 2026-09-27
"We automatically add the latest, most popular Hugging Face models to Model Garden."
- https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-garden/explore-models recorded 2026-09-27
"Model Garden is an AI/ML model library that helps you discover, test, customize, and deploy models and assets from Google and Google partners." "Deploy and serve open source model using Python SDK, CLI, REST API, or console"
Verified 2026-09-27