# jfalcou/eve

Expressive Vector Engine - SIMD in C++ Goes Brrrr

Repository: https://github.com/jfalcou/eve
Canonical: https://ross.abutalabs.com/products/jfalcou-eve
Homepage: https://jfalcou.github.io/eve/
Language: C++
License: BSL-1.0
License Family: other
Topics: cpp, simd, simd-library, simd-programming, simd-parallelism, hpc, cpp-library, sse2, ssse3, avx, avx2, altivec, neon, aarch64, arm-sve, arm-sve2, avx512
Last push: 2026-08-26T13:04:24+00:00

## Health v2 (maintenance only)
Score: 67/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 8, longevity 100
- inputs: {"age_days": 2792, "days_push": 7, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1360, forks 75 (observed 2026-08-28T04:04:29.945136+00:00)

## What it is
EVE is a C++20 header-only library providing a type-based, high-abstraction wrapper over SIMD instruction sets (SSE/AVX-512, NEON/SVE, AltiVec) for high-performance computing. It is a research-oriented reimplementation of the former Boost.SIMD library.

## Use cases
- write portable SIMD code in C++ without intrinsics
- vectorize numeric kernels across x86 and ARM
- accelerate math-heavy loops with AVX-512 or NEON
- write high-performance code that targets multiple SIMD ISAs
- replace hand-written SSE/AVX intrinsics with expressive C++20 abstractions

## When to choose
- you need portable explicit SIMD across Intel, ARM, and PowerPC
- you want high-level abstractions without sacrificing performance
- your codebase can use C++20
- you need wide mathematical function coverage over SIMD registers

## When to avoid
- you need GPU (GPGPU) vectorization
- you require ARM SVE with dynamic vector lengths
- you are stuck on pre-C++20 compilers
- you need a fully frozen, guaranteed-stable API before the 0.1 release

## Facets
- artifact type: library
- maturity: active
- function: simulation, math, developer-tools
- domain: performance, developer-tools, gpu-computing
- platform: cpp, cross-platform, windows
- tags: simd, hpc, vectorization, cpp20, header-only, avx512, neon, sve, altivec, algorithms, linux, macos

## Member repositories
- jfalcou/eve (main) score 67

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:29.945136+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 2026-08-30T04:41:39.323061+00:00, confidence not recorded.
  - readme: https://github.com/jfalcou/eve (fetched 2026-08-28T04:04:29.945136+00:00, sha 4676fe00e083)
  - homepage: https://jfalcou.github.io/eve/ (fetched 2026-08-29T11:59:25.771006+00:00, sha c4e99f4979f9)
- Data as of 2026-08-30T08:39:29.467469+00:00.