Llm Training Python Packages

skypilot

The AI Compute Platform for frontier teams. SkyPilot turns fragmented AI compute into one AI supercomputer, so frontier AI teams build custom intelligence faster.

liger-kernel

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

skypilot-nightly

The AI Compute Platform for frontier teams. SkyPilot turns fragmented AI compute into one AI supercomputer, so frontier AI teams build custom intelligence faster.

liger-kernel-nightly

Efficient Triton Kernels for LLM Training

trainy-skypilot-nightly

The AI Compute Platform for frontier teams. SkyPilot turns fragmented AI compute into one AI supercomputer, so frontier AI teams build custom intelligence faster.

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.

alignmenter

Check if your AI sounds like your brand, stays safe, and behaves consistently. Works with your custom GPTs, hosted APIs, and local models. Get detailed reports in minutes, not days.

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.

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

adafactor8bit

Memory-efficient PyTorch optimizers with Adaptive Log-Space quantization, fused CUDA kernels, and Adafactor, CAME, APOLLO support.

mlx-train-perf

Fused, logit-free linear-cross-entropy loss, RAM-fit planner, and benchmark harness for MLX fine-tuning on Apple Silicon

reef-infra

Infrastructure for continually self‑improving agents

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

candyllm

A simple, easy-to-use framework for HuggingFace and OpenAI text-generation models focused on explainability.

dlrover

DLRover: An Automatic Distributed Deep Learning System

autotransformers

A Python package for automatically training and comparing language models.

forgellm

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

nextai-prism

The AI Compute Platform for frontier teams. SkyPilot turns fragmented AI compute into one AI supercomputer, so frontier AI teams build custom intelligence faster.

ac-scheduler

The AI Compute Platform for frontier teams. SkyPilot turns fragmented AI compute into one AI supercomputer, so frontier AI teams build custom intelligence faster.

atorch

DLRover: An Automatic Distributed Deep Learning System