# Truss ## Docs - [Creating base images](https://baseten-philip-copy-0226.mintlify.app/contribute/base-images.md): Add a custom base image to Truss - [Contributing to Truss](https://baseten-philip-copy-0226.mintlify.app/contribute/contributing.md): Truss is MIT licensed and welcomes contributions - [Creating a PyPi release](https://baseten-philip-copy-0226.mintlify.app/contribute/release.md): Create a new version of Truss - [Setting up local development](https://baseten-philip-copy-0226.mintlify.app/contribute/setup.md): Get started working on Truss - [Getting Started](https://baseten-philip-copy-0226.mintlify.app/examples/01-getting-started-bert.md): Building your first Truss - [LLM](https://baseten-philip-copy-0226.mintlify.app/examples/02-llm.md): Building an LLM - [LLM with Streaming](https://baseten-philip-copy-0226.mintlify.app/examples/03-llm-with-streaming.md): Building an LLM with streaming output - [Text-to-image](https://baseten-philip-copy-0226.mintlify.app/examples/04-image-generation.md): Building a text-to-image model with SDXL - [Fast Cold Starts with Cached Weights](https://baseten-philip-copy-0226.mintlify.app/examples/06-high-performance-cached-weights.md): Deploy a language model, with the model weights cached at build time - [High Performance LLM with TGI](https://baseten-philip-copy-0226.mintlify.app/examples/07-high-performance-tgi.md): Deploy a language model with TGI - [High Performance LLM with vLLM](https://baseten-philip-copy-0226.mintlify.app/examples/08-high-performance-vllm.md): Deploy a language model with vLLM - [Private Hugging Face Model](https://baseten-philip-copy-0226.mintlify.app/examples/09-private-huggingface.md): Load a model that requires authentication with Hugging Face - [Model with system packages](https://baseten-philip-copy-0226.mintlify.app/examples/10-using-system-packages.md): Deploy a model with both Python and system dependencies - [How to use base images](https://baseten-philip-copy-0226.mintlify.app/guides/base-images.md): A guide to configuring a base image for your truss - [How to configure concurrency](https://baseten-philip-copy-0226.mintlify.app/guides/concurrency.md): A guide to setting concurrency for your model - [How to load model weights](https://baseten-philip-copy-0226.mintlify.app/guides/data-directory.md): Load model weights without Hugging Face or S3 - [How to use external packages](https://baseten-philip-copy-0226.mintlify.app/guides/external-packages.md): A guide on configuring your truss to use external packages - [How to cache model weights](https://baseten-philip-copy-0226.mintlify.app/guides/model-cache.md): Accelerate cold starts by caching your weights - [How to use secrets](https://baseten-philip-copy-0226.mintlify.app/guides/secrets.md): A guide to using secrets securely in your ML models - [High performance model serving with TGI](https://baseten-philip-copy-0226.mintlify.app/guides/tgi.md): A guide to using TGI for your model - [Learn model deployment](https://baseten-philip-copy-0226.mintlify.app/learn/intro.md): Learn to package and deploy ML models - [Step 0: Create a Truss](https://baseten-philip-copy-0226.mintlify.app/learn/model-serving/init.md): Set up your development environment and create a Truss - [Step 5: Implement model load](https://baseten-philip-copy-0226.mintlify.app/learn/model-serving/model-load.md): Load an ML model into your Truss - [Step 6: Implement model inference](https://baseten-philip-copy-0226.mintlify.app/learn/model-serving/model-predict.md): Add model inference and invoke the model server - [Step 2: Call the model server](https://baseten-philip-copy-0226.mintlify.app/learn/model-serving/predict.md): Call predict to test the model server - [Step 7: Publish your Truss](https://baseten-philip-copy-0226.mintlify.app/learn/model-serving/publish.md): Push your Truss to production-ready infrastructure - [Step 1: Spin up a model server](https://baseten-philip-copy-0226.mintlify.app/learn/model-serving/push.md): Configure a remote host and push the base Truss - [Step 4: Set Python requirements](https://baseten-philip-copy-0226.mintlify.app/learn/model-serving/requirements.md): Add required Python packages to the model server. - [Step 3: Enable live reload](https://baseten-philip-copy-0226.mintlify.app/learn/model-serving/watch.md): Patch the model server with local changes - [Quickstart](https://baseten-philip-copy-0226.mintlify.app/quickstart.md): Create, deploy, and invoke an ML model server in less than 5 minutes - [truss](https://baseten-philip-copy-0226.mintlify.app/reference/cli.md): The simplest way to serve models in production - [truss cleanup](https://baseten-philip-copy-0226.mintlify.app/reference/cli/cleanup.md): Clean up truss data. - [truss container](https://baseten-philip-copy-0226.mintlify.app/reference/cli/container.md): Subcommands for truss container. - [truss image](https://baseten-philip-copy-0226.mintlify.app/reference/cli/image.md): Subcommands for truss image. - [truss init](https://baseten-philip-copy-0226.mintlify.app/reference/cli/init.md): Create a new Truss. - [truss predict](https://baseten-philip-copy-0226.mintlify.app/reference/cli/predict.md): Invokes the packaged model. - [truss push](https://baseten-philip-copy-0226.mintlify.app/reference/cli/push.md): Pushes a truss to a TrussRemote. - [truss watch](https://baseten-philip-copy-0226.mintlify.app/reference/cli/watch.md): Seamless remote development with truss. - [Config options](https://baseten-philip-copy-0226.mintlify.app/reference/config.md): Set your model resources, dependencies, and more - [Deploy a Truss on Baseten](https://baseten-philip-copy-0226.mintlify.app/remotes/baseten.md): Use Baseten as a Truss remote - [User guide](https://baseten-philip-copy-0226.mintlify.app/usage.md): Get up and running quickly with Truss' developer workflow - [Welcome to Truss](https://baseten-philip-copy-0226.mintlify.app/welcome.md): The simplest way to serve AI/ML models in production ## Optional - [Baseten](https://www.baseten.co) - [Releases](https://github.com/basetenlabs/truss/releases/) - [Truss examples](https://github.com/basetenlabs/truss-examples/)