Llm Training Python Packages

skypilot

Run, manage, and scale AI workloads on any AI infrastructure. Use one system to access & manage all AI compute (Kubernetes, Slurm, 20+ clouds, on-prem).

liger-kernel

Efficient Triton Kernels for LLM Training

skypilot-nightly

Run, manage, and scale AI workloads on any AI infrastructure. Use one system to access & manage all AI compute (Kubernetes, Slurm, 20+ clouds, on-prem).

liger-kernel-nightly

Efficient Triton Kernels for LLM Training

yunchang

USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inference

trainy-skypilot-nightly

Run, manage, and scale AI workloads on any AI infrastructure. Use one system to access & manage all AI compute (Kubernetes, Slurm, 20+ clouds, on-prem).

ludwig

Low-code framework for building custom LLMs, neural networks, and other AI models

apollo-torch

APOLLO: SGD-like Memory, AdamW-level Performance; MLSys'25 Oustanding Paper Honorable Mention

ntloss

PyPI package for Number Token Loss (ICML 2025)

llm-magnet

the small distributed language model toolkit. fine-tune state-of-the-art LLMs anywhere, rapidly.

autotransformers

A Python package for automatically training and comparing language models.

atorch

DLRover: An Automatic Distributed Deep Learning System

nextai-prism

Prism: An intercloud broker for the clouds

candyllm

CandyLLM: Unified framework for HuggingFace and OpenAI Text-generation Models

obs2nlm

A simple tool to make a Google NotebookLM source from an Obsidian vault

dlrover

An Automatic Distributed Deep Learning Framework

alignmenter

Persona-aligned evaluation toolkit for auditing conversational AI authenticity, safety, and stability.

llm-trainer

🤖 Train your LLMs with ease and fun.

forgellm

A comprehensive toolkit for end-to-end continued pre-training, fine-tuning, monitoring, testing and publishing of language models with MLX-LM

rectokens

A tokenizer library for sequential recommendation systems

ac-scheduler

SkyPilot: Run AI on Any Infra — Unified, Faster, Cheaper. (Fork by Jason-AC)

sillm-mlx

Running and training LLMs on Apple Silicon via MLX

audiotoken

Audio tokenization, in the fastest way possible!

skypilot-mithril

Run, manage, and scale AI workloads on any AI infrastructure. Use one system to access & manage all AI compute (Kubernetes, Slurm, 20+ clouds, on-prem).