Hello, HN. I created L3Binance because I couldn't find any open-source engine that combined the high performance of Rust with real-time ML inference.

Key features:

     Latency-free AI: uses HuggingFace's Candle framework for inference within the Rust process (no Python overhead).
     Memory-optimized: uses Slab allocators for high-frequency agent management.

Detects spoofing, layering, and wash trading in real time.