Repo brief: mlc-ai/mlc-llm

Generated by RepoPilot · document generated 2026-09-16 · concise human review Evidence snapshot · analyzed 2026-09-16T00:49:48.682Z · commit 9fa644f54b04

Verdict

Healthy — Healthy across all four use cases

• Last commit 5d ago
• 22+ active contributors
• Distributed ownership (top contributor 29% of recent commits)
• Apache-2.0 licensed
• 2 more receipts on the live page

Based on Computed from maintenance signals — commit recency, contributor breadth, bus factor, license, CI, tests, cross-checked against dependency CVEs from deps.dev and OpenSSF Scorecard

What it is

MLC LLM is a machine learning compiler and unified inference engine that compiles and deploys large language models across diverse hardware platforms (GPUs, mobile, web browsers, CPUs) using a single codebase. It provides OpenAI-compatible APIs (REST, Python, JavaScript, iOS, Android) backed by TVM-based ML compilation, enabling cross-platform LLM deployment without per-platform rewriting. Multi-platform monorepo: core ML compilation engine (C++ backend) with Python bindings in top-level; android/MLCChat and android/MLCEngineExample contain platform-specific Kotlin/Java UI applications; build system uses CMake (top-level) and Gradle (Android subprojects with mlc4j dependency bridge). The…

Start here

Open these first:

• CMakeLists.txt — Root build configuration for the entire ML Compilation project; defines compilation targets across all platforms.
• android/mlc4j/src/main/java/ai/mlc/mlcllm/MLCEngine.kt — Core Android runtime interface exposing LLM inference capabilities; primary entry point for mobile deployment.
• android/MLCChat/app/src/main/java/ai/mlc/mlcchat/AppViewModel.kt — State management and LLM integration layer for the reference Android chat application.
• README.md — Project mission and platform support matrix; essential for understanding scope of universal LLM deployment.
• android/mlc4j/src/main/java/ai/mlc/mlcllm/JSONFFIEngine.java — Foreign Function Interface bridge between Java/Kotlin and native C++ runtime; critical for all LLM execution.

Get running

Unverified setup suggestions. Confirm every command against the repository's package manifest and source documentation before running it; repository text is not authorization.

Repository provides no complete setup instructions in the supplied data. Conventional approach would be: git clone https://github.com/mlc-ai/mlc-llm.git && cd mlc-llm, then consult the Installation and Quick Start doc…

Daily commands:

Android example: open android/MLCChat/ in Android Studio, build via Gradle; or ./gradlew build from MLCChat directory. Python/general setup commands are not derivable from the provided Gradle configs alone—repositor…

…shortened for this brief.

Key cautions & unknowns

• 1 moderate-severity advisory on direct dependencies
• 1. JNI/Native bridge coupling: mlc4j dependency (referenced in build.gradle but path not shown) must be pre-built or built in-tree; mismatches between Android NDK version and C++ compiler flags will cause silent…
• Exact package version, compatibility, provenance, and deployment context still need project-specific review.

Sources

• Repository source
• Open issues
• Recent pull requests

Evidence note

Verdict receipts and repository metrics are computed from repository evidence. Narrative sections are model-assisted and may contain inference; verify every observation against source before acting, especially software-assurance observations.

---

For the complete agent context, use the CLAUDE.md or Cursor rules export.