mixture-of-experts

12 projets partagent ce topic GitHub

Les traits relient les membres réellement apparentés entre eux. La taille des points suit les étoiles.

🧬 Membres

A 2.78-trillion-parameter Kimi K3 running inference on a single CPU in 8.24 GB of RAM. Portable C99: no BLAS,…

Optimizing inference proxy for LLMs

mixtral-offloading

Run Mixtral-8x7B models in Colab or consumer desktops

Codebase for Aria - an Open Multimodal Native MoE

From scratch implementation of a sparse mixture of experts language model inspired by Andrej Karpathy's…

Chinese-Mixtral

中文Mixtral混合专家大模型（Chinese Mixtral MoE LLMs）

Krasis is a Hybrid LLM runtime which focuses on efficient running of larger models on consumer grade VRAM…

A library for easily merging multiple LLM experts, and efficiently train the merged LLM.

PyTorch library for cost-effective, fast and easy serving of MoE models.

soft-moe-pytorch

Implementation of Soft MoE, proposed by Brain's Vision team, in Pytorch

Project_Chronos

⚡ Zero-Stall MoE Inference via Lookahead Prediction & Async DMA Prefetching. Optimized for SSD I/O with…

Toy-scale unified multimodal model experiments — encoder-free understanding & generation with…

🔗 Familles voisines

llm-inference 65 inference-engine 12 transformer 59 moe 13 quantization 36 lora 47 from-scratch 7 simd 6

Mesuré à partir des topics GitHub communs aux deux projets, pondérés par leur rareté.