Mixture of experts
7 updates on Mixture of experts.
Liquid AI releases LFM2.5-24B-A2B, a 2B-active MoE for local agents
Liquid AI reports 184 tok/s on an Apple M5 Max for its 24B-parameter mixture-of-experts model, which activates 2.45B parameters per token.
Edge0 releases MoE expert-offloading framework, demos 35B model on iPhone
Edge0 streams mixture-of-experts weights from storage. Its 35B tier reports 2.9 GiB peak memory on a Mac mini M4 Pro; a launch post shows a 35B model on an iPhone.
Qualcomm teases next-gen Hexagon NPU ahead of Snapdragon Summit 2026
Qualcomm has teased its next-generation Hexagon NPU — featuring a transformer-focused Element Accelerator and 50% more shared memory — built for on-device agentic AI ahead of the …
Meta AI builds MobileMoE, 5.3B parameters with 0.9B active per token
Meta AI trained three on-device mixture-of-experts models that store 1.3B to 5.3B parameters and run 272M to 922M of them per token.
D2MoE picks a bit-width per token, 1.39 times the throughput at up to 53 percent less memory
A MobiCom 2025 paper routes every token to an expert and a bit-width, reporting up to 1.39 times the throughput of EdgeMoE at up to 53 percent less memory.
CoMiGS splits on-device fine-tuning into shared generalists and private specialists
EPFL researchers split LoRA adapters into shared generalists and device-local specialists, and report 47.19 test perplexity against 54.38 for training alone.
Ai2 releases OLMoE, 7B parameters with 1B active per token
Ai2's mixture-of-experts model holds 6.9B parameters but runs 1.3B per token, and its iOS app runs it offline on an iPhone 15 Pro or newer.