Axolotl AI ORG

Go ahead and axolotl questions

A framework for fine-tuning open-source AI language models like Llama, Mistral, and Qwen using a single YAML config file, handling distributed multi-GPU training, LoRA, and quantization without writing training code from scratch.

ARC Relay — WebSocket relay server for agent remote control by Axolotl AI

A fast, local, and secure approach for training LLMs for coding tasks using GRPO with WebAssembly and interpreter feedback.

No description.

Axolotl Plugin integrating with Nous Research's Atropos

No description.

Differential transformers implementation and axolotl plugin

A public showcase of configs, WandB logs, and the resulting models built with Axolotl OSS

No description.

Context Parallelism utilities for Training Language Models

Dion optimizer algorithm

No description.

Efficient triton implementation of Native Sparse Attention.

Benchmarking LLMs with Challenging Tasks from Real Users

No description.

Exact speculative decoding on Apple Silicon, powered by MLX.

Official inference library for pre-processing of Mistral models

LPGL licensed contribs to axolotl

Ring attention implementation with flash attention

🤗 The largest hub of ready-to-use datasets for ML models with fast, easy-to-use and efficient data manipulation tools

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

🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.