Unsloth

This release focuses heavily on stabilizing and fixing issues related to Gemma 4 training and quant formats, alongside significant stability and feature enhancements in Unsloth Studio, including the addition of speculative decoding.

This release introduces full support for the new Gemma 4 model family, including low-RAM variants, alongside substantial improvements to tool calling accuracy, stability, and web search functionality. Numerous fixes were also applied across the Unsloth Studio environment and underlying build processes, particularly concerning llama.cpp integration and dependency pinning.

This release significantly improves tool call accuracy and reliability across models, introduces features like custom folders and preliminary multi-GPU support, and resolves numerous stability and installation issues across platforms.

This release delivers significant performance improvements, making inference 20-30% faster, and introduces auto-detection for models from sources like LM Studio and Hugging Face. Numerous bug fixes address stability, installation logging, and Studio UI responsiveness.

This release focuses heavily on Unsloth Studio improvements, including major performance gains via pre-compiled binaries, enhanced tool calling, and robust fixes for Windows and Colab environments. Installation sizes have been significantly reduced, and UX features like persistent settings and multi-file uploads have been added.

This release introduces install-ready Unsloth Studio llama.cpp bundles corresponding to commit b8475.

This release provides install-ready Unsloth Studio llama.cpp bundles updated for revision b8457.

This release introduces Unsloth Studio (Beta), a new web UI for training and running LLMs, alongside significant feature additions like Mixtral support, ROCM compatibility, and various stability and security fixes across the platform.

This release introduces Unsloth Studio, a new open-source web UI for training and running LLMs locally, alongside significant performance improvements and broad model support including Mixtral and ROCM.

This release introduces major performance enhancements, including 12x faster MoE training and ultra-long context RL capabilities, alongside support for several new state-of-the-art models. Numerous bug fixes address stability across various architectures and dependencies.

This release introduces massive performance gains with 3x faster training via new Triton kernels and enables 500K context length fine-tuning. It also adds support for Transformers v5, preliminary multi-GPU training, and several new model guides.

This release introduces major performance enhancements with FP8 Reinforcement Learning support and significant VRAM reductions across the board. It also adds support for new models like DeepSeek-OCR and Qwen3-VL, alongside improved Docker integration.

This release introduces major platform support, including Docker images and Blackwell/DGX hardware compatibility, alongside significant new features like Quantization-Aware Training (QAT) and extensive RL environment utilities.

This release introduces significant performance gains and new capabilities for gpt-oss Reinforcement Learning, alongside support for new models like DeepSeek-V3.1-Terminus and Magistral 1.2. Several bug fixes were implemented, including resolving issues with BERT and QAT + LoRA fast path.

This release introduces major performance enhancements and new capabilities for Vision models in Reinforcement Learning (RL), alongside the new 'Standby' feature for memory-efficient training. Numerous bug fixes and improvements were also integrated across various components, including Intel/ROCm support and serialization workflows.

This release introduces Unsloth Flex Attention for gpt-oss training, drastically improving context length, VRAM efficiency, and speed. Numerous bug fixes and support for new models/features like QAT + LoRA are also included.

This release introduces broad support for the new gpt-oss model, enabling low-VRAM fine-tuning, alongside significant algorithmic updates that improve performance across all models. It also adds support for Qwen3 models and expands compatibility to include newer NVIDIA hardware like RTX 50 series and Blackwell GPUs.

This release focuses heavily on stability, VRAM reduction (10-25% less), and broad model compatibility, including full fixes for Gemma 3N Vision and support for new models like Devstral 1.1 and MedGemma.

This release introduces major new capabilities including support for multimodal Gemma 3n models and Text-to-Speech fine-tuning, alongside new quantization methods (Dynamic 2.0 GGUFs) and support for DeepSeek-R1-0528 and Magistral-24B.

This release introduces official support for Qwen3 models, including fine-tuning capabilities for the 30B MoE variant. Numerous bug fixes address compatibility issues, quantization errors, and improve overall stability.

The March release introduces full support for finetuning Gemma 3 models and significantly expands model compatibility, including Mixtral and vision models, alongside preliminary support for 8bit and full finetuning. This version also brings Windows support and removes the compilation requirement for GGUF exports.

This release introduces GRPO, achieving up to 90% memory reduction during training, alongside various bug fixes and updates to support Llama 3.1 8B training.

This release introduces major support for GRPO training, enabling LoRA/QLoRA for GRPO across various models, and integrates fast inference via vLLM for significant throughput gains. Numerous bug fixes address issues with Gemma 2, Mistral mapping, and general stability.

This release introduces full support for the Phi-4 model, including fixes for tokenization and chat templates, alongside significant bug fixes for gradient accumulation, vision models, and performance regressions. It also brings Windows support and performance improvements via updated Xformers.