QVAC

QVAC vs MLC LLM

QVAC

Open-source ecosystem for local-first peer-to-peer AI on every platform.

MLC LLM

A machine learning compiler and deployment engine for language models, mobile included.

Key differences

MLC LLM and QVAC both run models on iOS, Android, desktop and server, both expose a JavaScript API, and both carry Apache 2.0 licences. MLC LLM compiles each model ahead of time through the TVM stack, which is where its throughput on tuned hardware comes from. QVAC loads GGUF at runtime through its Fabric engine.

Task coverage differs. MLC LLM deploys language models. QVAC covers twelve task types behind one API, including speech recognition, speech synthesis, OCR, translation and image generation.

The workflows differ accordingly. MLC LLM requires a per-target compile step before a model ships. QVAC reads a model from a local file, an HTTPS URL or a peer at runtime, so adding or replacing one is a configuration change. Ahead-of-time compilation trades flexibility for tuning, and runtime loading trades tuning for flexibility.

This page compares MLC LLM v0.20.0, released 19 August 2026, against QVAC 0.18.2, meaning the SDK together with the Fabric inference engine at v10297.1.1. Every row was checked against the project's own documentation and release notes on 4 September 2026. Both projects move quickly, so check the current release before you make a decision on either one.

Feature matrix

Feature

QVAC

MLC LLM

PLATFORMS

macOS

Yes
Yes

Windows

Yes
Yes

Linux

Yes
Yes

Android

Yes
Yes

iOS

Yes
Yes

AI TASKS

Text generation

Yes
Yes

Transcription

Yes
No

Translation

Yes
No

Image generation

Yes
No

OCR

Yes
No

Text-to-speech

Yes
No

RUNTIME SUPPORT

Node.js

Yes
Yes

Bare

Yes
No

Expo

Yes
No

HTTP server

Yes
Yes

CLI

Yes
Yes

P2P

Peer discovery

Yes
No

Inference delegation

Yes
No

Encrypted transport

Yes
No

MOBILE SUPPORT

On-device inference

Yes
Yes

LoRA fine-tuning on mobile

Yes
No

Mobile SDK

Yes
Yes

LICENSING

License

Apache 2.0

Apache 2.0

Open weights tooling

Yes
Yes

When to choose QVAC

Your application needs speech, OCR, translation or images.

You want to add or swap a model without recompiling it.

You are building in JavaScript, TypeScript or Python.

You need on-device fine-tuning or peer-to-peer delegation.

When to choose MLC LLM

You want ahead-of-time compilation tuned to one target.

You are already working in the TVM ecosystem.

Language models are the only workload.

You want the same engine in a browser over WebGPU.

Compare

Ready to build with QVAC?

One SDK, every platform, no rent. Grab it and ship your own local-first AI.

npm install @qvac/sdk
Get the SDKRead the docs