Unsloth: Releases and Overview
Library for fast, low-VRAM fine-tuning (LoRA / QLoRA). Also known for GGUF export and its Dynamic 2.0 quantized releases.
Category: Quantization, model formats and fine-tuning. Part of our engine and tool tracker.
At a Glance
| Item | Value |
|---|---|
| Repository | unslothai/unsloth |
| License | Apache-2.0 |
| Main language | Python |
| GitHub stars | 76,273 (as of 2026-09-17) |
| Latest release | v0.1.811-beta (2026-09-19) |
| Install / docs | official documentation |
The project describes itself as: “Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4, MiniMax-H3, Gemma 4, FLUX and more.”
Release History
Compiled by Local Model Watch from the project’s GitHub releases. Pre-releases are omitted. Where we wrote an article about a release, it is linked in the last column; smaller releases are tracked here only.
| Version | Released | Release notes | Our article |
|---|---|---|---|
| v0.1.811-beta (Docker + Multi User + AMD Support) | 2026-09-19 | GitHub | Unsloth v0.1.811-beta Released with AMD & NVIDIA Docker Images |
| v0.1.810-beta (Docker + Multi User + AMD Support) | 2026-09-17 | GitHub | Unsloth v0.1.810-beta Released with Multi-User and AMD Support |
| Windows-ARM64 (Windows ARM64 Binaries) | 2026-09-16 | GitHub | — |
| v0.1.808-beta (Large Performance Gains + Fixes) | 2026-09-10 | GitHub | — |
| v0.1.807-beta (Large Perf Improvements + Fixes) | 2026-09-08 | GitHub | — |
| v0.1.806-beta (2x Faster Qwen3.8-Flash + GLM-5.3-Flash MTP) | 2026-09-02 | GitHub | — |
| v0.1.805-beta (2x Faster Qwen3.8-Flash + GLM-5.3-Flash MTP) | 2026-09-02 | GitHub | — |
| v0.1.804-beta (Qwen3.8-Flash-Next + GLM-5.3-Flash) | 2026-08-27 | GitHub | — |
| v0.1.803-beta (Bug Fixes + Auto compaction + LAN Remote Access) | 2026-08-25 | GitHub | — |
| v0.1.802-beta (Bug Fixes + Auto compaction + LAN Remote Access) | 2026-08-25 | GitHub | — |
| v0.1.801-beta (Auto compaction (preview) + LAN Remote Access) | 2026-08-21 | GitHub | — |
| v0.1.800-beta (Qwen3.8-27B) | 2026-08-14 | GitHub | — |
| v0.1.702-beta | 2026-08-14 | GitHub | — |
| v0.1.701-beta (Introducing Unsloth Desktop 🦥) | 2026-08-12 | GitHub | — |
| v0.1.70-beta (Introducing Unsloth Desktop 🦥) | 2026-08-12 | GitHub | — |
| v0.1.62-beta | 2026-08-11 | GitHub | — |
| v0.1.61-beta (Meta Muse Glimmer) | 2026-08-10 | GitHub | — |
| v0.1.60-beta (Meta Muse Glimmer) | 2026-08-10 | GitHub | — |
| v0.1.527-beta | 2026-08-10 | GitHub | — |
| v0.1.526-beta (DSpark + DeepSeek-V4 Flash 0731) | 2026-08-05 | GitHub | — |
| v0.1.512-beta ( Kimi K3 + DeepSeek-V4 Flash 0731 + Deep Research + Parallel Chat) | 2026-07-30 | GitHub | — |
| v0.1.501-beta (Introducing AMD support) | 2026-07-20 | GitHub | — |
| v0.1.49-beta (New Personalization and Language Support!) | 2026-07-16 | GitHub | — |
| v0.1.481-beta (DeepSeek-V4 + NVFP4 Exporting) | 2026-07-07 | GitHub | — |
| v0.1.471-beta (GLM 5.2 + Model Hub + 3x longer contexts) | 2026-06-19 | GitHub | — |
| v0.1.464-beta (DiffusionGemma + Gemma 4 MTP) | 2026-06-12 | GitHub | — |
| v0.1.451-beta (Gemma 4 MTP + Bug Fixes) | 2026-06-11 | GitHub | — |
| v0.1.44-beta (Gemma 4 12B, New UI, MCP, Projects) | 2026-06-03 | GitHub | — |
| v0.1.43-beta (CUDA 13.3, Windows, Mac update) | 2026-05-31 | GitHub | — |
| v0.1.42-beta (An Update before Revamp!) | 2026-05-26 | GitHub | — |
| v0.1.41-beta (MTP + Studio fixes) | 2026-05-19 | GitHub | — |
| v0.1.405-beta (Qwen3.6 MTP and API / Connections) | 2026-05-18 | GitHub | — |
| v0.1.39-beta (New Unsloth API Inference Endpoint) | 2026-05-05 | GitHub | — |
| v0.1.37-beta (New UI Redesign + Qwen3.6) | 2026-04-23 | GitHub | — |
| v0.1.36-beta (Gemma 4 Fixes) | 2026-04-08 | GitHub | — |
| v0.1.35-beta (Google – Gemma 4 now in Unsloth!) | 2026-04-03 | GitHub | — |
| v0.1.3-beta (+50% tool call accuracy) | 2026-04-01 | GitHub | — |
| v0.1.25-beta (New Important Updates!) | 2026-03-28 | GitHub | — |
| v0.1.2-beta (First Release post Unsloth Studio!) | 2026-03-26 | GitHub | — |
| b8475 | 2026-03-23 | GitHub | — |
| b8457 | 2026-03-21 | GitHub | — |
| v0.1.0-beta (Introducing Unsloth Studio (Beta)!) | 2026-03-18 | GitHub | — |
| February-2026 (12x Faster MoE Training + Embedding support!) | 2026-02-11 | GitHub | — |
| December-2025 (December Release + 3x Faster Training) | 2025-12-19 | GitHub | — |
| November-2025 (November Release + FP8 Training!) | 2025-11-26 | GitHub | — |
| October-2025 (October Release + Unsloth Docker!) | 2025-10-27 | GitHub | — |
| September-2025-v3 (gpt-oss Reinforcement Learning + Auto Kernel Notebook) | 2025-09-27 | GitHub | — |
| September-2025-v2 (Vision Reinforcement Learning + Memory Efficient RL) | 2025-09-17 | GitHub | — |
| August-2025-v2 (Unsloth Flex Attention + Long context gpt-oss Training) | 2025-08-29 | GitHub | — |
| August-2025 (gpt-oss Fine-tuning) | 2025-08-09 | GitHub | — |
| July-2025 (Less VRAM + bug fixes) | 2025-07-10 | GitHub | — |
| June-2025 (Gemma 3n + Text-to-speech (TTS)) | 2025-06-27 | GitHub | — |
| May-2025 (Qwen3) | 2025-05-03 | GitHub | — |
| 2025-03 (Gemma 3 + FFT Support) | 2025-03-15 | GitHub | — |
| 2025-02-v2 (Long Context GRPO) | 2025-02-21 | GitHub | — |
| 2025-02 (GRPO, vLLM) | 2025-02-06 | GitHub | — |
| 2025-01 (Phi-4 & Bug Fixes) | 2025-01-11 | GitHub | — |
| December-2024 ( Llama 3.3 + Dynamic 4bit Quants) | 2024-12-04 | GitHub | — |
| November-2024 (Vision finetuning) | 2024-11-22 | GitHub | — |
| October-2024 (Gradient Accumulation Fix) | 2024-10-16 | GitHub | — |
Articles on Local Model Watch
- 2026-09-19 Unsloth v0.1.811-beta Released with AMD & NVIDIA Docker Images
- 2026-09-18 Unsloth v0.1.810-beta Released with Multi-User and AMD Support
Last updated 2026-09-19 (JST). Facts above come from the GitHub API; the one-paragraph summary is written by the site’s editors.