OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Infrastructure & Serving

Truss

Truss is an open-source Python CLI and model-packaging tool that bundles machine-learning serving logic, weights and dependencies, configures container and GPU requirements, launches training jobs, and deploys models to Baseten.

Independent OpenViST project profile, based on linked sources.

Truss — OpenViST project profile in AI Infrastructure & Serving.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Truss when a machine-learning model needs a repeatable Python packaging and deployment workflow for Baseten, including configuration of dependencies, containers, GPUs and supported inference engines.

Who it is for

  • ML engineers deploying models through Baseten
  • Python developers packaging custom inference logic
  • Teams serving open-source models with GPU-backed infrastructure
  • Builders using supported Python and inference frameworks

Documented uses

  • Package model-serving code, weights and dependencies
  • Define model, engine, hardware and inference settings in a project configuration
  • Containerize and deploy models to Baseten
  • Launch training jobs and connect resulting artifacts to serving workflows
  • Create production inference endpoints for supported models
Before you choose

Requirements & limits

The documented quickstart requires a Baseten account, the uv package manager and Baseten authentication through browser login or an API key. Truss is distributed under the MIT license.

  • Baseten is the deployment platform used in the documented quickstart.
  • The workflow supports both engine-based configuration and custom Python model code.
  • The Qwen example uses an L4 GPU and two builder GPUs for FP8 TensorRT-LLM compilation.
  • The tool supports dependency, container and GPU configuration as part of the model package.
  • Development-mode live reload is not supported for the documented TensorRT-LLM deployment.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

COMPARE OPTIONS

Compare options for Model inference and serving

These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.

ProjectDocumented purposePlatformAccess and requirements
TrussPackage model-serving code, weights and dependenciesCLI, Cloud, Docker, Self-hostedThe documented quickstart requires a Baseten account, the uv package manager and Baseten authentication through browser login or an API key. Truss is distributed under the MIT license.
Cloud AI SDKAdapt and run documented model recipes for computer-vision and NLP inference.Cloud, ServerThe GitHub repository is public. Its public visibility does not establish license terms.
TurboquantReplace a model’s existing inference cache with a compressed cache.Cloud, Linux, macOS, Self-hostedThe project describes itself as open source. The README documents installation through pip or from a source checkout.
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
3 October 2026
Original website title
basetenlabs/truss · Official homepage title · retained in its source language
How we review sourcesSuggest a correction