llmlip-rs: Rust MLIP inference engine runs 6000-atom sims 17.7x faster than PyTorch MACE
CatAstro_Piyush · x · 2026-10-31
Mizukami Group open-sourced llmlip-rs, a Rust inference engine for machine-learning interatomic potentials with no PyTorch or JAX dependency at runtime, shipping a Rust library, CLI, C ABI, and LAMMPS/OpenMM plugins under Apache-2.0.
Benchmarks on a 16-thread Apple Silicon CPU with a 6000-atom periodic water box: PyTorch MACE FP64 takes 13.09 s per step vs 0.74 s for llmlip-rs FP32 — 17.7x faster. Attached charts also show 6000-atom silicon on a Mac GPU via Metal at 196 ms per step, 5.1x faster than its CPU path.
More from Infra
- Nvidia weighs cosigning loan for Nscale's West Virginia AI data center — TechCrunch AI · 2026-10-31
- Northflank dunked on Railway's breach, then got hacked the same way 9 days later — aniketmaurya · 2026-10-31
- Dev says Neon is already chasing millisecond infra branching, wants it built into alchemy — samgoodwin89 · 2026-10-31
- Big AI labs all push compute efficiency: Trainium power savings, small models, cache compression — WillfullyWorkable · 2026-10-31
- Running 12 agents on one RTX 3090 lifts total throughput from 30 to 110 tok/s — Mchanger · 2026-10-31
- Analyst turns ultra-bull on Avicena after 6 hours of technical diligence — BenBajarin · 2026-10-31