Unsloth
Desktop and web app for running and fine-tuning local LLMs and diffusion models
What it is
Unsloth runs open models such as Qwen, Gemma, DeepSeek and FLUX locally, and fine-tunes LLMs, diffusion, TTS and embedding models with LoRA, QLoRA, full fine-tuning and reinforcement learning methods like GRPO and DPO while using less GPU memory. It ships as a desktop app, a web UI, and a Python library, and can export models to GGUF and other formats or serve them through an OpenAI-compatible API. Developers and researchers use it to customize models on their own hardware and to point coding agents like Claude Code or Codex at local models.