Go 1.27 adds portable SIMD API modeled on Highway, unlocking AI inference speedups
arpit_bhayani · x · 2026-09-26
The official Go Blog details the experimental SIMD APIs shipping in Go 1.26 and 1.27.
- Context: SIMD crunches vectors in a single instruction (e.g., 8 float64 additions at once); until now Go required hand-written assembly, so most software left CPU capability unused
- Roadmap: Go 1.26 added arch-specific APIs for amd64 (archsimd package); Go 1.27 added arm64 NEON and wasm, then introduced a fully portable, size-agnostic simd package loosely based on C++ Highway — write once, near-asm performance with emulation fallback
- Real-world use: Go's Green Tea garbage collector already uses SIMD to speed up memory scanning
- Significant for AI inference, cryptography, and data processing workloads
Related event: Go 1.27 Introduces Experimental Cross-Platform SIMD API(2 posts)→
More from Infra
- Three years building CPUs and spiking nets: one engineer's path to AI accelerators — k7agar · 2026-11-11
- CoreWeave: 2027 GPU renewals could add ~$4B annualized revenue, analyst skeptical — zephyr_z9 · 2026-11-11
- Running DeepSeek 4.1 locally on an $800 rig with a used RTX 3090 at 2-3 tok/s — LankyGuitar6528 · 2026-11-11
- Mark Says George Hotz Was Prescient About Nvidia's Situation — mark_k · 2026-11-11
- Amazon Prime 360 uses AI to rebuild NBA plays in 3D for live, any-angle replays — import_jmr · 2026-11-11
- Altman says OpenAI aims to replace Nvidia in the medium to long term — zephyr_z9 · 2026-11-11