DeepEP
DeepEP is an open-source communication library built by DeepSeek for distributed training and inference, written first for Mixture-of-Experts (MoE) models.
Mixture of Experts
Explore Mixture of Experts through related topics and the articles other pages reference most.
Explore articles
Browse subtopics: AI Infrastructure
Articles that also belong to these categories. Counts cover all of Mixture of Experts.
All articles (63)Large Language Models (55)Open Source AI (43)AI Models (41)Chinese AI (34)Multimodal AI (11)AI Agents (3)AI Infrastructure (3)Model Architecture (3)Reasoning Models (3)Transformer Models (3)AI Code Generation (2)Diffusion Models (2)Meta AI (2)NVIDIA (2)AI Companies (1)AI Inference (1)Generative AI (1)Google (1)Google DeepMind (1)Healthcare AI (1)Natural Language Processing (1)OpenAI (1)Research Organizations (1)Training & Optimization (1)Video Generation (1)
Showing 1-3 of 3 articles
Expert Parallelism
Expert Parallelism (EP) is a model-parallelism strategy specific to Mixture of Experts (MoE) neural networks in which the individual expert sub-networks (typically feed-forward blocks) are sharded across…
MoonEP
MoonEP is an open-source expert-parallel communication library for mixture-of-experts models, released by Moonshot AI on July 27, 2026 under the MIT License.