Moe Python Packages

sglang

SGLang is a high-performance serving framework for large language models and multimodal models.

vllm

A high-throughput and memory-efficient inference and serving engine for LLMs

flashinfer-python

FlashInfer: Kernel Library for LLM Serving

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.

flashinfer-cubin

FlashInfer: Kernel Library for LLM Serving

llamafactory

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

sglang-kernel

SGLang is a high-performance serving framework for large language models and multimodal models.

sgl-kernel

SGLang is a high-performance serving framework for large language models and multimodal models.

ms-swift

Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.6, DeepSeek-V4, GLM-5.1, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM4.5v, Gemma4, Llava, Phi4, ...) (AAAI 2025).

vllm-tpu

A high-throughput and memory-efficient inference and serving engine for LLMs

vllm-cpu-nightly

A high-throughput and memory-efficient inference and serving engine for LLMs

tensorrt-llm

TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and supports state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs. TensorRT LLM also contains components to create Python and C++ runtimes that orchestrate the inference execution in a performant way.

sawyer-core

Distributed MoE inference network — the load is split, friends help.

sglang-kt

SGLang is a high-performance serving framework for large language models and multimodal models.

awex

A high-performance RL training-inference weight synchronization framework, designed to enable second-level parameter updates from training to inference in RL workflows

switch-transformers

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

llmtuner

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

abliterix

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

uccl

UCCL: Ultra and Unified CCL

thekaveh-nnx

Lightweight PyTorch toolkit for training, fine-tuning, and exporting modern neural nets. FFN + GNN + decoder-only LM + diffusion + JEPA + MoE; PEFT (LoRA/DoRA/IA3/Prefix/Prompt), PTQ/QAT, pruning, surgery; ONNX/GGUF/Ollama/HuggingFace Hub interop. Dataclass-configured runs with fluent builders, automatic checkpointing, and Plotly viz.

lsglang

SGLang is a fast serving framework for large language models and vision language models.

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.

lazyllm-llamafactory

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

ai-dynamo-vllm

A high-throughput and memory-efficient inference and serving engine for LLMs