Pinned Loading

1. Air.rs Air.rs Public

   ⚡ High-performance, zero-config, hardware-agnostic LLM inference engine in pure Rust with direct zero-F32 quantization & "infinite" context.

   Rust 10

2. OmniToken OmniToken Public

   ⚡ Blazing-fast, universal tokenization engine in Rust. Unifies BPE, WordPiece, and Unigram into a single trie-walker automaton with AVX2 SIMD acceleration. Achieves 3.02+ GB/s multi-core throughput…

   Rust 1

3. SunayHegde2006.github.io SunayHegde2006.github.io Public