simdutf/simdutf

Health v2 · maintenance only

Adoption not part of the score

1938 stars · 149 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

simdutf is a high-performance C++ library for Unicode validation, transcoding (UTF-8, UTF-16, UTF-32, Latin1, ASCII) and WHATWG-compliant Base64 encode/decode, accelerated with SIMD instructions. It achieves billions of characters per second and is used in production by Node.js, WebKit, Chromium, Bun, and Cloudflare Workers.

Use cases

• validate utf-8 strings at high speed
• convert between utf-8 and utf-16
• transcode unicode text with simd acceleration
• decode and encode base64 fast
• check if byte sequences are valid unicode
• fast ascii validation in c++
• convert latin1 to utf-8

When to choose

• you need maximum-throughput unicode validation or transcoding in C/C++
• you are building a runtime, parser, or server that processes large volumes of text
• you need WHATWG-compliant base64 at multi-GB/s speeds
• you want a dependency-free, allocation-free, noexcept library with a C API option

When to avoid

• you need higher-level text processing like collation, normalization, or case mapping (use ICU)
• you work in a managed language with adequate built-in encoding support and no performance bottleneck
• you need locale-aware string operations

Facets

library · maturity stable

serialization parser developer-tools developer-tools performance cpp cross-platform windows simd unicode utf8 utf16 utf32 base64 transcoding validation avx2 avx-512 neon risc-v high-performance algorithms natural-language-processing linux macos gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="simdutf/simdutf")