[[["易于理解","easyToUnderstand","thumb-up"],["解决了我的问题","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["很难理解","hardToUnderstand","thumb-down"],["信息或示例代码不正确","incorrectInformationOrSampleCode","thumb-down"],["没有我需要的信息/示例","missingTheInformationSamplesINeed","thumb-down"],["翻译问题","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["最后更新时间 (UTC):2025-08-29。"],[],[],null,["# Overview of Model Garden\n\nModel Garden is an AI/ML model library that helps you discover, test,\ncustomize, and deploy models and assets from Google and Google partners.\n\nAdvantages of Model Garden\n--------------------------\n\nWhen you're working with AI models, Model Garden provides the following\nadvantages:\n\n- Available models are all grouped in a single location\n- Model Garden provides a consistent deployment pattern for different types of models\n- Model Garden provides built-in integration with other parts of Vertex AI such as model tuning, evaluation, and serving\n- Serving generative AI models can be difficult---Vertex AI handles model deployment and serving for you\n\nExplore models\n--------------\n\nTo view the list of available Vertex AI and open source foundation,\ntunable, and task-specific models, go to the Model Garden page in the\nGoogle Cloud console.\n\n[Go to Model Garden](https://console.cloud.google.com/vertex-ai/model-garden)\n\nThe model categories available in Model Garden are:\n\nTo filter models in the filter pane, specify the following:\n\n- **Tasks**: Click the task that you want the model to perform.\n- **Model collections**: Click to choose models that are managed by Google, partners, or you.\n- **Providers**: Click the provider of the model.\n- **Features**: Click the features that you want in the model.\n\nTo learn more about each model, click its model card.\n\nFor a list of models available in Model Garden, see\n[Models available in Model Garden](/vertex-ai/generative-ai/docs/model-garden/available-models).\n\nModel security scanning\n-----------------------\n\nGoogle does thorough testing and benchmarking on the serving and tuning\ncontainers that we provide. Active vulnerability scanning is also applied to\ncontainer artifacts.\n\nThird-party models from featured partners undergo model checkpoint scans to\nensure authenticity. Third-party models from HuggingFace Hub are scanned\ndirectly by HuggingFace and their\n[third-party scanner](https://huggingface.co/docs/hub/en/security-protectai)\nfor malware, pickle files, Keras Lambda layers, and secrets. Models deemed\nunsafe from these scans are flagged by HuggingFace and blocked from deployment\nin Model Garden. Models deemed suspicious or those that have the\nability to potentially execute remote code are indicated in\nModel Garden but can still be deployed. We recommend you perform a\nthorough review of any suspicious model before deploying it\nwithin Model Garden.\n\nPricing\n-------\n\nFor the open source models in Model Garden, you are charged for use of\nfollowing on Vertex AI:\n\n- **Model tuning** : You are charged for the compute resources used at the same rate as custom training. See [custom training pricing](/vertex-ai/pricing#custom-trained_models).\n- **Model deployment** : You are charged for the compute resources used to deploy the model to an endpoint. See [predictions pricing](/vertex-ai/pricing#prediction-prices).\n- **Colab Enterprise** : See [Colab Enterprise pricing](/colab/pricing).\n\nControl access to specific models\n---------------------------------\n\nYou can set a [Model Garden organization\npolicy](/vertex-ai/generative-ai/docs/control-model-access) at the organization, folder, or\nproject level to control access to specific models in Model Garden. For\nexample, you can allow access to specific models that you've vetted and deny\naccess to all others.\n\nLearn more about Model Garden\n-----------------------------\n\nFor more information about the deployment options and customizations that you\ncan do with models in Model Garden, view the resources in the\nfollowing sections, which include links to tutorials, references, notebooks, and\nYouTube videos.\n\n### Deploy and serve\n\nLearn more about customizing deployments and advance serving features.\n\n- [Deploy and serve open source model using Python SDK, CLI, REST API, or console](/vertex-ai/generative-ai/docs/model-garden/use-models#deploy_a_model)\n - [Developer blog: Introducing the new Vertex AI Model Garden CLI and SDK](https://www.googlecloudcommunity.com/gc/Community-Blogs/Introducing-the-new-Vertex-AI-Model-Garden-CLI-and-SDK/ba-p/888386)\n - [Deploy open models by using the SDK tutorial notebook](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_deployment_tutorial.ipynb)\n - [Get started with Vertex AI Model Garden SDK notebook](https://github.com/GoogleCloudPlatform/generative-ai/blob/main/open-models/get_started_with_model_garden_sdk.ipynb)\n- [Deploying and fine-tuning Gemma 3 in Model Garden YouTube video](https://youtu.be/pC2DhFJQocY?si=xxFsBH7A0XgKMOSh)\n- [Deploying Gemma and making predictions](/vertex-ai/generative-ai/docs/model-garden/deploy-and-inference-tutorial)\n- [Serve open models with a Hex-LLM container on Cloud\n TPUs](/vertex-ai/generative-ai/docs/open-models/use-hex-llm)\n- [Deploying Llama models by using Hex-LLM tutorial notebook](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_hexllm_deep_dive_tutorial.ipynb)\n- [Use prefix caching and speculative decoding with\n Hex-LLM or vLLM tutorial notebook](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_advanced_features.ipynb)\n- [Use vLLM to serve text-only and multimodel language models on Cloud GPUs](/vertex-ai/generative-ai/docs/open-models/vllm/use-vllm)\n - [Text-only models tutorial notebook](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_vllm_text_only_tutorial.ipynb)\n - [Multimodal models tutorial notebook](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_vllm_multimodal_tutorial.ipynb)\n- [Use xDiT GPU serving container for image and video generation](/vertex-ai/generative-ai/docs/open-models/xdit)\n- [Serving Gemma 2 with multiple LoRA adapters with HuggingFace DLC for PyTorch inference tutorial on Medium](https://medium.com/google-cloud/open-models-on-vertex-ai-with-hugging-face-serving-multiple-lora-adapters-on-vertex-ai-e3ceae7b717c)\n- [Use custom handles to serve PaliGemma for image captioning with HuggingFace DLC for PyTorch inference tutorial on LinkedIn](https://www.linkedin.com/posts/ivan-nardini_vertexai-huggingface-deeplearning-activity-7269824291238526976-oxuM?utm_source=li_share&utm_content=feedcontent&utm_medium=g_dt_web&utm_campaign=copy)\n- [Deploy and serve a model that uses Spot VMs or a Compute Engine reservation tutorial notebook](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_reservations_spotvm.ipynb)\n- [Deploy and serve a HuggingFace model](/vertex-ai/generative-ai/docs/open-models/use-hugging-face-models)\n\n#### Container compliance\n\nModel Garden offers the following FedRAMP high compliant\ncontainers for model serving.\n\n### Tuning\n\nLearn more about tuning models to tailor responses for specific use cases.\n\n- [Workbench fine-tuning tutorial notebook](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_llama3_1_finetuning_with_workbench.ipynb)\n- [Fine-tuning and evaluation tutorial notebook](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_finetuning_tutorial.ipynb)\n- [Deploying and fine-tuning Gemma 3 in Model Garden YouTube video](https://youtu.be/pC2DhFJQocY?si=xxFsBH7A0XgKMOSh)\n\n### Evaluation\n\nLearn more about assessing model responses with Vertex AI\n\n- [Evaluate Gemma 2 with the generative AI evaluation service YouTube video](https://youtu.be/AUSunZXC2rg?si=SEag-u-a9KtA7gK1)\n\n### Additional resources\n\n- [Model and user journey-specific Model Garden notebooks](https://github.com/GoogleCloudPlatform/vertex-ai-samples/tree/main/notebooks/community/model_garden)\n- [Vertex AI open model serving, fine-tuning and evaluation notebooks](https://github.com/GoogleCloudPlatform/generative-ai/tree/main/open-models)"]]