Mixture Of Experts Python Packages

nvidia-cudnn-frontend

cuDNN Frontend is NVIDIA's modern, open-source entry point to the cuDNN library and a growing collection of high-performance open-source kernels.

deepspeed

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

egobox

Efficient global optimization toolbox in Rust: bayesian optimization, mixture of gaussian processes, sampling methods

mixture-of-experts

A Pytorch implementation of Sparsely-Gated Mixture of Experts, for massively increasing the parameter count of language models

smt

SMT: The Surrogate Modeling Toolbox

experts4bit-qlora

Fine-tune and serve large Mixture-of-Experts models on one GPU, with 4-bit experts in GPU memory, RAM or SSD.

grouped-nf4-gemm

4-bit GPU kernels that compute directly on packed Mixture-of-Experts weights, for training and inference.

mlx-kquant

Native K-quant support for MLX, with a quantization and fine-tuning toolchain for Apple Silicon

hivemind

Decentralized deep learning in PyTorch. Built to train models on thousands of volunteers across the world.

optillm

Optimizing inference proxy for LLMs

loggetta

Plan and run Mixture-of-Experts fine-tuning on one GPU: check the machine, fit the model, and keep a report of each run.

neural-trees

sklearn-compatible PyTorch implementations of Soft Decision Trees, HMoE, and classifier comparison tests (5×2cv F-test). pip install neural-trees

krasis

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

mlx-flash

Run AI models too large for your Mac's memory — at near-full speed. Intelligent expert caching, speculative execution, and 15+ research techniques for MoE inference on Apple Silicon.

abliterix

Automated alignment adjustment for LLMs — direct steering, LoRA, and MoE expert-granular abliteration, optimized via multi-objective Optuna TPE.

aether-quant

Dynamic multi-asset trading system (equities, crypto, bonds, futures, options, Forex) on Lean + PyTorch. Mixture-of-Experts, causal-TCN, 3D topology, rank-driven long/short book, and continuous controlled retraining.

mila-llm

A C++23 library for local LLMs at the metal — inference and training, built from explicit neural-network components you can read and understand.

sparklab

Run frontier open-weight models privately on NVIDIA DGX Spark.

st-moe-pytorch

Implementation of ST-Moe, the latest incarnation of MoE after years of research at Brain, in Pytorch

peer-pytorch

Pytorch implementation of the PEER block from the paper, Mixture of A Million Experts, by Xu Owen He at Deepmind

dantinox

DantinoX: A modular, memory-efficient Transformer implementation in JAX/Flax NNX. Includes Sparse MoE, GQA, Sliding Window Attention, Gradient Accumulation and Checkpointing

switch-transformers

Implementation of Switch Transformers from the paper: "Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity"

dissenter

A PyPI package implementing a multi-LLM ensemble debate framework. Routes prompts across Claude, GPT, Gemini, Ollama, and any LiteLLM-compatible provider; surfaces disagreements across models and synthesizes a consensus answer.

bigrig

Run large Mixture-of-Experts models on an Apple Silicon Mac that does not have room for them, and know what it cost