gpt-oss

This release focuses heavily on enhancing the Unsloth Studio experience with a major UI/UX overhaul, new chat features, and expanded model support, alongside significant stability and hardware compatibility fixes, particularly for 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 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 improved handling for builtin and unsolicited tool calls within the gpt-oss component and provides numerous pre-compiled binaries for various operating systems and hardware configurations.

This release introduces support for 'reasoning_format = none' for gpt-oss models and provides updated binary distributions for macOS, Linux, Windows, and openEuler across various hardware and acceleration backends.

This release replaces the `wrap_for_generation` utility with a new prefix convenience function and resolves a bug affecting gpt-oss.

This release introduces a new Web Search API for real-time information retrieval and expands the new engine's capabilities to support Qwen3 architectures and multi-regex pretokenizers.

This release adds CUDA 13 support, introduces a dimensions field for embeddings, and improves memory estimation and app UI. It also removes support for loading split vision models in the Ollama engine.

This release enables flash attention by default for gpt-oss models and improves their overall loading performance.

This release introduces significant memory management improvements for GPU scheduling and multi-GPU setups, alongside performance optimizations for gpt-oss models and reduced installation sizes.

This release fixes a VRAM leak in gpt-oss during multi-device execution and improves Windows stability by statically linking C++ libraries.

This patch release focuses on stability improvements for gpt-oss, specifically fixing crashes related to KV cache quantization and a missing variable definition.