ANEMLL (pronounced like "animal") is an open-source project focused on accelerating the porting of Large Language Models (LLMs) to tensor processors, starting with the Apple Neural Engine (ANE).
The goal is to provide a fully open-source pipeline from model conversion to inference for common LLM architectures running on ANE.
This enables seamless integration and on-device inference for low-power applications on edge devices, ensuring maximum privacy and security.
This is critical for autonomous applications, where models run directly on the device without requiring an internet connection.
This test model is exclusively for the QWEN3 architecture converted for CoreML, released before the official launch of the ANEMLL repository and minimal documentation. It is intended for early adopters only who requested an early release.
Requirements
macOS Sequoia with Apple Neural Engine and 8GB RAM or more
CoreML Tools and HuggingFace Transformers libraries
Python 3.9
chat.py provides a sample inference script. chat_full.py provides a sample inference script with history and conversation management.
Installation
Download the model from Hugging Face:
bash
1# Install required tools2pip install huggingface_hub
34# Install Git LFS (Large File Support)5# macOS with Homebrew:6brew install git-lfs
7# Or Ubuntu/Debian:8# sudo apt-get install git-lfs910# Initialize Git LFS11git lfs install1213# Clone the repository with model files14git clone https://huggingface.co/anemll/anemll-Qwen3-0.6B-MLX-dequantized-ctx512_0.3.0
Extract model files:
bash
1# Navigate to cloned directory2cd anemll-Qwen3-0.6B-ctx512_0.3.0
34# Pull LFS files (model weights)5git lfs pull
67# Extract CoreML model files8find. -type f -name "*.zip" -exec unzip{}\;
Note: The first time the model loads, macOS will take some time to place it on the device.
Subsequent loads will be instantaneous.
Use Ctrl-D to exit, Ctrl-C to interrupt inference.
More Info
Please check following links for later updates:
