AIPython★ 349⑂ 31 forksApache-2.0
lithos-ai/lithos-metal
High-Performance LLM Kernels for Apple
Provides optimized inference kernels and a serving layer for large language models on Apple silicon, enabling fast, low-latency execution in Python.
// KEY FEATURES
- Accelerates LLM inference with custom Metal kernels on M-series chips.
- Provides a lightweight Python serving API for deploying models locally.
- Supports batching and dynamic shape handling for throughput optimization.
New repositories every 30 minutes
REDDYX AI scans GitHub 24/7 and ships the best AI/ML/Web3 projects to Telegram.
Join on Telegram