#mixture-of-experts Repositories
Discover 78 active open source GitHub repositories tagged with #mixture-of-experts, accumulating 392,641 total stars (averaging 5,033 stars per project). Leading this topic is deepseek-ai/DeepSeek-V3 with 104,417 stars.
78 repositories total
DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.
A 2.78-trillion-parameter Kimi K3 running inference on a single CPU in 8.24 GB of RAM. Portable C99: no BLAS, no framework, no GPU.
ใIn-Depth Understanding of AI Infrastructure: Quantitative Analysis and System Designใ (by Li Bojie), an open-source book manuscript: Starting from hardware constraints and model architectures, it quantitatively derives the design of LLM inference and training systems. Includes the full text, PDF, accompanying computational tools, and experiments.
Production-tested AI infrastructure tools for efficient AGI development and community-driven innovation
DeepSeek-Coder-V2: Breaking the Barrier of Closed-Source Models in Code Intelligence
DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding
fastllm is a high-performance large language model inference library with no backend dependencies, helping you run inference efficiently on most devices!
DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model
Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models
Qwen3.8 is the large language model series developed by the Qwen team at Alibaba Group.
Miles is an enterprise-facing reinforcement learning framework for LLM and VLM post-training, forked from and co-evolving with slime.
Run GLM-5.3 (744B), DeepSeek-V4.1-Flash (748B), MiniMax-M3.1-Flash (428B), GLM-5.3-Flash (321B), and Hy4-Flash (220B) on your own GPUs โ starting from 8 GB of VRAM, with NVIDIA and AMD support, one dashboard, and an OpenAI- and Anthropic-compatible API.
Building Large Models from Scratch: A Complete Practice from Pretraining to RLHF
Decentralized deep learning in PyTorch. Built to train models on thousands of volunteers across the world.
Monkey (LMM): Image Resolution and Text Label Are Important Things for Large Multi-modal Models (CVPR 2024 Highlight)
DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models
Re-implementation of pi0 vision-language-action (VLA) model from Physical Intelligence
Spiking Brain-inspired Large Models, integrating hybrid efficient attention, MoE modules and spike encoding into its architecture
Local inference that behaves like a frontier API: Qwen3.8-Flash-Next on AMD Strix Halo plus small NPU models, over the OpenAI and Anthropic APIs on one port.
MiMo-V2-Flash: Efficient Reasoning, Coding, and Agentic Foundation Model
Zero-GPU CPU inference for 30Bโ284B MoE models (Qwen3.8 176.5B cold-cache certified in 28.26/48.35 GiB Cgroup v2, DeepSeek-V4 284B, Nemotron-30B) + RouteCrack 2.0.
Run large AI models on your own PC: Qwen3.8-Flash-Next on Strata's engine, and GLM 5.3 Flash, DeepSeek V4 Flash, LFM2.5, Qwen3.6, Ornith 1.5, Gemma 4, Tiel Coder, and MiniCPM-V on a second engine built for PolyStrata. One-click installation for Windows and Linux, with an OpenAI/Anthropic API on localhost.
Run models larger than your RAM or VRAM by streaming MoE experts from SSD. Built on llama.cpp.
PyTorch Re-Implementation of "The Sparsely-Gated Mixture-of-Experts Layer" by Noam Shazeer et al. https://arxiv.org/abs/1701.06538
Kimi-VL: Mixture-of-Experts Vision-Language Model for Multimodal Reasoning, Long-Context Understanding, and Strong Agent Capabilities
Frequently Asked Questions
What is the most popular #mixture-of-experts repository on GitHub?
The top ranked open source #mixture-of-experts project on GitHub is deepseek-ai/DeepSeek-V3 with 104,417 stars.
Which programming languages are most used in #mixture-of-experts repositories?
Among the top ranked #mixture-of-experts projects on GitHub, the most common programming languages are Python, C++, Jupyter Notebook, C.
How is the #mixture-of-experts repository ranking calculated?
We track 78 active open source repositories tagged with #mixture-of-experts (500+ stars), totaling 392,641 stars with an average of 5,033 stars per repository. Rankings update daily from the GitHub API.