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

Utensor

Utensor is an Apache-2.0 C++ machine-learning inference framework with a runtime library and an offline tool that translates TensorFlow models into C++11 source for embedded targets.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Utensor — 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 uTensor when the task is to bring a TensorFlow-trained model into an embedded C++ application for local inference, or to explore the repository's TinyML tutorials.

Who it is for

  • Embedded C++ developers adding local machine-learning inference to microcontroller or IoT projects.
  • TensorFlow users preparing trained models for inference on Arm-focused hardware.

Documented uses

  • Translate a TensorFlow model into C++ source for embedded inference.
  • Build tutorial executables to explore uTensor's error handling and custom-operator examples.
  • Integrate inference into an embedded C++ application and configure tensor metadata and temporary-data allocators.
Before you choose

Requirements & limits

The repository is public and declares the Apache-2.0 license. Its tutorial build instructions require CMake and make.

  • The documented workflow uses TensorFlow for model construction and training, then generates C++11 inference source.
  • The repository describes an embedded runtime rather than a hosted inference API.
  • The tutorial build produces executables, while board-specific deployment instructions are not included in the documented build steps.

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

Source-based practical guide

Build the uTensor tutorial executables

Configure and build the repository's tutorial examples.

Before you start

  • A checkout of the uTensor repository
  • CMake available on the system
  • A build environment with make

Steps

  1. Create a build directory in the repository and enter it.

  2. Run CMake with the PACKAGE_TUTORIALS option enabled.

  3. Run make to build the tutorials.

  4. Find the resulting executables under build/tutorials/ and consult each tutorial directory's README for its usage instructions.

Expected outcome

Tutorial executables are placed in build/tutorials/.

Guide limitations

  • This procedure builds tutorial examples; it does not document deployment to a specific embedded board.
Documentation 1 ↗
Evidence and transparency

Sources & review

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