What is Unsloth used for?

Unsloth is an open-source app and framework for running and fine-tuning LLMs and diffusion models locally, with GGUF and MLX support and an OpenAI-compatible API. It is a good fit for fine-tuning open models on a single GPU, running private local AI with tool calling and serving local models to coding agents.

Is Unsloth open source?

Yes. Unsloth is open source under the Apache-2.0 licence. Its source code is on GitHub at unslothai/unsloth and is written mainly in Python.

Yes. Unsloth is open source, so the software itself is free to use.

Can I self-host Unsloth?

Yes. Unsloth can be self-hosted on your own server or infrastructure; there is no official hosted version.

What is Unsloth an alternative to?

Unsloth is an open-source alternative to Together AI. Other open-source alternatives to Together AI include LocalAI, vLLM and SGLang.

Is Unsloth actively maintained?

Yes. The most recent commit to Unsloth was on 2 October 2026, and the latest release is v0.1.902-beta, published on 1 October 2026. The project has 77k stars on GitHub.

Unsloth: open-source alternative to Together AI

About Unsloth

Unsloth is an open-source framework with a desktop app for running and training AI models on your own hardware. The same tool covers inference of local language, image, audio and embedding models as well as fine-tuning them, and it runs on Windows, Linux (including WSL) and macOS, with support for NVIDIA, AMD and Intel GPUs, CPUs and a Vulkan backend.

On the running side it loads GGUF and MLX models, serves them through an OpenAI-compatible API, and lets coding agents such as Claude Code and Codex use local models with tool calling and code execution. It also offers private web search, deep research and retrieval, plus access from other devices over a LAN or a secure remote link. For training it supports LoRA, QLoRA, full fine-tuning, pretraining and reinforcement learning methods such as GRPO and DPO, with export to formats including GGUF.

The project claims fine-tuning that is roughly twice as fast with about 70 percent less VRAM, which was its original draw. Unsloth is Apache-2.0 licensed, can be installed from native packages or a Docker image, and has notebooks and documentation for getting started.

Key features

Run GGUF and MLX models locally
Fine-tune LLMs with LoRA and QLoRA
Reinforcement learning with GRPO and DPO
OpenAI-compatible API for local models
Native desktop app and Docker image
Export trained models to GGUF

Good fit for

→Fine-tuning open models on a single GPU
→Running private local AI with tool calling
→Serving local models to coding agents

Built with: Python
Tags: llm; fine-tuning; local-ai; gguf; lora; reinforcement-learning; desktop-app; python; self-hosted

Unsloth: questions and answers

What is Unsloth used for?: Unsloth is an open-source app and framework for running and fine-tuning LLMs and diffusion models locally, with GGUF and MLX support and an OpenAI-compatible API. It is a good fit for fine-tuning open models on a single GPU, running private local AI with tool calling and serving local models to coding agents.
Is Unsloth open source?: Yes. Unsloth is open source under the Apache-2.0 licence. Its source code is on GitHub at unslothai/unsloth and is written mainly in Python.
Is Unsloth free?: Yes. Unsloth is open source, so the software itself is free to use.
Can I self-host Unsloth?: Yes. Unsloth can be self-hosted on your own server or infrastructure; there is no official hosted version.
What is Unsloth an alternative to?: Unsloth is an open-source alternative to Together AI. Other open-source alternatives to Together AI include LocalAI, vLLM and SGLang.
Is Unsloth actively maintained?: Yes. The most recent commit to Unsloth was on 2 October 2026, and the latest release is v0.1.902-beta, published on 1 October 2026. The project has 77k stars on GitHub.