cd/entity/OLMoE-1B-7B-0924· home› entities› OLMoE-1B-7B-0924
grep -l @olmoe-1b-7b-0924 /news/*.json | wc -l → 1

OLMoE-1B-7B-0924

mentions 1 type Organization feed RSS

// recent coverage 1 mentions

22:37
2026-09-01
github.com
machine-learning

Moe expert offloading on a 2-core Celeron with 2.7GB RAM

A developer's on-hardware test shows that prefetching mixture-of-experts (MoE) expert weights from disk ahead of matmuls can improve token generation speed on a severely resource-constrained machine, …

// co-occurs with top 4 entities
Intel Celeron N4000 1 llama.cpp 1 Qwen2.5-0.5B-Instruct 1 ChromeOS Crostini 1
// topics top 3 topics
machine learning 1 large language models 1 ai infrastructure 1