Apache TVM is the backbone of OctoAI
Apache TVM is an open-source deep-learning compiler framework that empowers engineers to optimize and run computations efficiently, on any hardware backend.
Open-source for optimal performance
Simplified deployment
Enables compilation of deep learning models in PyTorch, TensorFlow, and many more into minimum deployable modules on diverse hardware backends.
Automatic optimization
Provides the infrastructure to automatically generate and optimize tensor operators on more backends with better performance.
TVM Unity
Empowering developers to optimize machine learning models and universally deploy them on any hardware backend.
Meet the fastest growing open-source ML community
Started as a research project at the SAMPL group of Paul G. Allen School of Computer Science & Engineering, University of Washington, TVM is now in incubation at The Apache Software Foundation (ASF). With 900+ contributors and 6,000+ members using the framework for optimizations — including Amazon, Facebook, Arm and Qualcomm, and more — it is quickly becoming the standard for deep learning compilation. Also, TVM is the engine behind its sister project MLC-LLM, which provides a universal environment to deploy LLMs on a broad range of devices.
See who’s putting TVM to work
TVM enables two kinds of optimizations
Computational graph optimization performs tasks such as high-level operator fusion, layout transformation, and memory management.
Tensor operator optimization and code generation layer that optimizes tensor operators.
It's easy to get involved
Learn from the tutorials
Compile PyTorch, TensorFlow, or CoreML models, and deploy pre-trained models on Android.
See the current TVM roadmap
Discover what's next with TVM — auto scheduling, quantization, auto tensorization, and more.
Join the discussion forums
Tap into the collective knowledge and expertise of over over 3,000 users (and counting!).
Design and developer guide
Jump into the core and help us build the future of MLOps.
Your choice of models on our SaaS or in your environment
Run any model or checkpoint on our efficient, reliable, and customizable API endpoints. Sign up and start building in minutes.