# simd-everywhere/simde

Implementations of SIMD instruction sets for systems which don't natively support them.

Repository: https://github.com/simd-everywhere/simde
Canonical: https://ross.abutalabs.com/products/simde
Homepage: https://simd-everywhere.github.io/blog/
Language: C
License: MIT
License Family: permissive
Topics: simd-intrinsics, sse, neon, arm, avx, simd, sse2, sse3, ssse3, sse41, sse42, avx2, avx512, fma, gfni, mmx, altivec, powerpc, arm64, vectorization
Last push: 2026-08-26T15:02:23+00:00

## Health v2 (maintenance only)
Score: 78/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 39, longevity 100
- inputs: {"age_days": 3445, "days_push": 7, "days_rel": 192, "gap_med": 192.5, "n_releases_24m": 3}
- flags: prerelease_only
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3092, forks 320 (observed 2026-08-28T04:07:42.382217+00:00)

## What it is
SIMDe is a header-only C/C++ library providing portable implementations of SIMD intrinsics (SSE, AVX, NEON, AltiVec, etc.) on hardware that doesn't natively support them. It enables porting SIMD-optimized code across architectures with no performance penalty when native support exists.

## Use cases
- port SSE/AVX code to ARM without rewriting
- run NEON code on x86 for development without an emulator
- get a cross-architecture port working quickly before optimizing
- write code against ISA extensions not available on local hardware
- keep SIMD code portable across x86, ARM, and PowerPC

## When to choose
- you need to port existing SIMD intrinsics code to another architecture
- you want a gradual migration path rather than a full rewrite
- you need to develop/test SIMD code for hardware you don't have access to

## When to avoid
- you only target a single architecture and can use native intrinsics directly
- you want a higher-level abstraction like a portable vector library rather than intrinsic-level compatibility
- maximum performance on unsupported hardware is critical, since emulated paths are slower

## Facets
- artifact type: library
- maturity: active
- function: graphics, simulation, developer-tools
- domain: cross-platform, developer-tools, performance
- platform: cross-platform, cpp, c, windows
- tags: simd, intrinsics, header-only, sse, avx, neon, altivec, portability, vectorization, algorithms, linux, macos

## Member repositories
- simd-everywhere/simde (main) score 78

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:42.382217+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:46:23.931811+00:00, confidence not recorded.
  - readme: https://github.com/simd-everywhere/simde (fetched 2026-08-28T04:07:42.382217+00:00, sha bcb5e98ef362)
  - homepage: https://simd-everywhere.github.io/blog/ (fetched 2026-08-29T09:42:33.887212+00:00, sha 92f517a68497)
- Data as of 2026-08-30T08:39:29.467469+00:00.