unsloth
SHA-256Unsloth is the first desktop app to run and train AI models locally. It supports a wide range of models including Qwen3.8, Kimi K3, MiniMax-H3, Gemma 4, DeepSeek-V4, and FLUX. Unlike most local model tools that focus only on inference, Unsloth adds a full training stack: fine-tune LLMs, diffusion, TTS, and embedding models 2× faster with 70% less VRAM. It also supports LoRA, QLoRA, RL, GRPO, DPO, and FP8. The app includes built-in Agents & Tools integration (Claude Code, Codex, MCP), private search and RAG, image/video diffusion, and audio support. With hardware support for CPU, NVIDIA, AMD, Intel, macOS, and multi-GPU setups, plus remote access via Cloudflare, Unsloth is a comprehensive local AI studio. It can export to GGUF, NVFP4, FP8 and offers an OpenAI-compatible API for easy deployment.
Smart Download
Download Download Version
v0.1.701-beta · 44.4 MB
A free, open-source desktop app to run and train LLMs and diffusion models locally, with up to 2× faster fine-tuning and 70% less VRAM.
Core Features
- Run and train LLMs, diffusion models, audio, embedding, and more
- 2× faster fine-tuning with 70% less VRAM compared to conventional methods
- Supports LoRA, QLoRA, full fine-tuning, RL, GRPO, DPO, and FP8
- Works with agents like Claude Code and Codex, plus private search and RAG
- Hardware support for CPU, NVIDIA, AMD, Intel, macOS, and multi-GPU
What It Can't Do
- •Even with fast fine-tuning, large training tasks still need sufficient RAM and VRAM; check hardware requirements first
- •First launch may require downloading model files, so ensure a stable internet connection
- •The Vulkan backend only accelerates GGUF inference, not training; training still requires a PyTorch or MLX backend
Use Cases
- Fine-tune open-source models (e.g., Qwen, DeepSeek, Llama) locally for custom tasks
- Run a local model hub with agent integration, private knowledge base, and offline inference
Detailed Introduction
Unsloth is the first desktop app to run and train AI models locally. It supports a wide range of models including Qwen3.8, Kimi K3, MiniMax-H3, Gemma 4, DeepSeek-V4, and FLUX. Unlike most local model tools that focus only on inference, Unsloth adds a full training stack: fine-tune LLMs, diffusion, TTS, and embedding models 2× faster with 70% less VRAM. It also supports LoRA, QLoRA, RL, GRPO, DPO, and FP8. The app includes built-in Agents & Tools integration (Claude Code, Codex, MCP), private search and RAG, image/video diffusion, and audio support. With hardware support for CPU, NVIDIA, AMD, Intel, macOS, and multi-GPU setups, plus remote access via Cloudflare, Unsloth is a comprehensive local AI studio. It can export to GGUF, NVFP4, FP8 and offers an OpenAI-compatible API for easy deployment.
Tags
Getting Started
Download installer
Click the button above to download the installer for your system
Install the software
Install the appropriate package for your distro (dpkg / rpm / AppImage)
Download the installer for your operating system (Windows/macOS/Linux)
Double-click the installer and follow the setup wizard
Launch the Unsloth desktop app, select or download a model, and start using it
- Download the installer for your operating system (Windows/macOS/Linux)
- Double-click the installer and follow the setup wizard
- Launch the Unsloth desktop app, select or download a model, and start using it
SHA-256 checksum verified
Checksum extracted from GitHub official Release page
SHA256 Checksum
3f9fe4489d724d746909e082693a067f6b7e7803d99c5c95efe1ff604edf88e5This checksum is extracted from the GitHub Release page. Verify file integrity after download.
All SHA-256 checksums on this platform are extracted from the project's official GitHub Release page, without any modification. You can independently verify them on the GitHub Releases page.
Open Source Transparency
View GitHub SourceUninstall Info
On Windows, use Settings > Apps to uninstall; on macOS, drag the app to the Trash; on Linux, use your package manager (e.g., apt remove unsloth).
No Extra Dependencies
Ready to use after download. No additional runtime required.
Similar Projects
ragflow
RAGFlow is a leading open-source RAG engine that fuses deep document understanding with agentic workflows, offering precise, citation-grounded answers. Unlike LangChain's DIY approach, RAGFlow provides a turnkey, enterprise-ready platform with a visual interface and template-based chunking, reducing hallucinations and setup complexity.
LocalAI
LocalAI is the open-source AI engine to run any model - LLMs, vision, voice, image, video - on any hardware. No GPU required. Drop-in API compatibility with OpenAI, Anthropic, and ElevenLabs.
ollama
Ollama lets you download, run, and manage large language models locally. One command, multiple platforms, endless possibilities.