Up to 200x Faster Inner Products and Vector Similarity — for Python, JavaScript, Rust, C, and Swift, supporting f64, f32, f16 real & complex, i8, and binary vectors using SIMD for both x86 AVX2 & AVX-512 and Arm NEON & SVE 📐
USearch already uses SimSIMD for hardware-accelerated distance functions. That, however, exceptionally complicates backward-compatible compilation and testing. Instead, for Python users, we can wrap SimSIMD functions into CPython capsules to pass into USearch from the usearch.index.Index.
USearch already uses SimSIMD for hardware-accelerated distance functions. That, however, exceptionally complicates backward-compatible compilation and testing. Instead, for Python users, we can wrap SimSIMD functions into CPython capsules to pass into USearch from the
usearch.index.Index.