MLC LLM
Project record
Maintained by MLC team (MLC open community)16, CMU Catalyst (lists MLC LLM among its own research projects)8
MLC LLM is an open-source machine learning compiler and deployment engine for large language models. Compiled models run on MLCEngine, which offers an OpenAI-compatible API through a REST server and Python, JavaScript, iOS, and Android interfaces, with GPU backends including CUDA, ROCm, Vulkan, Metal, WebGPU, and OpenCL.1
- Repository: GitHub repository (external site: github.com)
- Documentation: Documentation (external site: llm.mlc.ai)
- Documentation: Installation (external site: llm.mlc.ai)
- License: License (Apache 2.0) (external site: raw.githubusercontent.com)
Availability and license
Overall availability
Source code on GitHub under the Apache License 2.0. The installation guide distributes nightly pre-release pip wheels, recommends installing them into a conda environment, and also documents building from source.124
Availability is separate from permission: read the license before using or redistributing.
Component reuse rights
- weights
- Unknown — no complete fact-level rights review
- code
- Unknown — no complete fact-level rights review
- data
- Unknown — no complete fact-level rights review
- documentation
- Unknown — no complete fact-level rights review
No complete system-rights review is recorded for this release.
Public materials checklist
| Item | Status | Notes and evidence |
|---|---|---|
| Source codeIs the source code publicly readable? | Public | Published on GitHub under the Apache License 2.0.12 |
| DocumentationIs user documentation published? | Public | Documentation covers installation, a quick start, and deployment to the web, iOS, Android, REST, Python, and the command line.3 |
| InstallationAre installation instructions or packages publicly available? | Public | Nightly-built pip wheels from the MLC wheel index or building from source; the guide recommends installing into a conda environment.4 |
| Supported platformsAre supported operating systems or hardware documented? | Public | The README gives a support matrix covering Linux, Windows, macOS, web browsers (WebGPU and WASM), iOS/iPadOS, and Android, with AMD, NVIDIA, Apple, Intel, Adreno, and Mali GPUs.14 |
| Release statusAre versioned releases published? | Partial | The installation guide distributes nightly pre-release builds. The repository's tags include v0.20.0 (July 7, 2026), whose commit message describes it as a stable release; the installation guide does not document a packaged stable release.45 |
What it is useful for
Deploying language models natively on desktop and server GPUs, mobile devices, and in web browsers from a shared engine and compiler.1
Organization context
U.S. eligibility
Eligible · basis: U.S.-governed project
Carnegie Mellon University's Catalyst research group lists MLC LLM as one of its own research projects, which documents a U.S. university lab as a maintaining entity. The project credits the "MLC team" and describes MLC as an open community; every organization the MLC site lists as supporting and contributing (NSF, CMU, Catalyst, Purdue, NVIDIA, Amazon, and Google) is U.S.-based. MLC's 2023 announcement says the project was started by members of CMU Catalyst, UW SAMPL, Shanghai Jiao Tong University, OctoML, and the MLC community; that multi-institution origin is recorded here. Eligibility rests on the documented CMU lab, not on contributors' affiliations or nationality.8617
Sources
This listing is not an endorsement, a safety assessment, or a federal approval.