microsoft/onnxruntime-genai v0.12.0

v0.12.0

Repository: microsoft/onnxruntime-genai

Tag: v0.12.0

Published: 2026-02-13T17:38:22Z

Prerelease: no

Release notes:

What's Changed

• Update versions after making 0.11.0 branch by @kunal-vaishnavi in https://github.com/microsoft/onnxruntime-genai/pull/1867
• Fix guidance usage in continuous decoding by @kunal-vaishnavi in https://github.com/microsoft/onnxruntime-genai/pull/1870
• Fix HelloPhi C# example by @kunal-vaishnavi in https://github.com/microsoft/onnxruntime-genai/pull/1871
• Fix regex by @apsonawane in https://github.com/microsoft/onnxruntime-genai/pull/1875
• Update extensions commit by @apsonawane in https://github.com/microsoft/onnxruntime-genai/pull/1874
• Revert removal of eps_without_if_support by @xiaofeihan1 in https://github.com/microsoft/onnxruntime-genai/pull/1878
• Fix condition for NPU by @apsonawane in https://github.com/microsoft/onnxruntime-genai/pull/1880
• Model builder refactoring by @tianleiwu in https://github.com/microsoft/onnxruntime-genai/pull/1862
• Add lintrunner to format code by @tianleiwu in https://github.com/microsoft/onnxruntime-genai/pull/1884
• Remove empty submodule leftover. by @xkszltl in https://github.com/microsoft/onnxruntime-genai/pull/1883
• Fix build for lack of RTLD_DI_ORIGIN support by @jaeyoonjung in https://github.com/microsoft/onnxruntime-genai/pull/1888
• Enable graph capture for webgpu by @qjia7 in https://github.com/microsoft/onnxruntime-genai/pull/1848
• Generic shared emb_tokens/lm_head implementation by @jixiongdeng in https://github.com/microsoft/onnxruntime-genai/pull/1885
• Fix bug in Squeeze for getting the value of total_seq_len by @Honry in https://github.com/microsoft/onnxruntime-genai/pull/1886
• Extra_options disable_qkv_fusion to untie qkv_projs from upstream choice by @jixiongdeng in https://github.com/microsoft/onnxruntime-genai/pull/1893
• Fix mac pipeline by @apsonawane in https://github.com/microsoft/onnxruntime-genai/pull/1904
• whisper: Support a variant of the whisper pipeline where encoder / decoder are stateful. by @RyanMetcalfeInt8 in https://github.com/microsoft/onnxruntime-genai/pull/1857
• Add model builder for Qwen2_5_VLTextModel by @tianleiwu in https://github.com/microsoft/onnxruntime-genai/pull/1882
• Integrate FARA-7B model by @apsonawane in https://github.com/microsoft/onnxruntime-genai/pull/1902
• Fix gpt-oss model export by @apsonawane in https://github.com/microsoft/onnxruntime-genai/pull/1861
• OpenVINO: Add support for model caching via 'cache_dir' provider option by @RyanMetcalfeInt8 in https://github.com/microsoft/onnxruntime-genai/pull/1900
• WinML - Remove the inclusive Microsoft.WindowsAppSDK.ML range check by @chrisdMSFT in https://github.com/microsoft/onnxruntime-genai/pull/1907
• Run the model in text mode by @apsonawane in https://github.com/microsoft/onnxruntime-genai/pull/1908
• Update extensions commit by @apsonawane in https://github.com/microsoft/onnxruntime-genai/pull/1914
• Fix gpt-oss export by @apsonawane in https://github.com/microsoft/onnxruntime-genai/pull/1915
• Support Olive new uint8 quantization format by @xiaoyu-work in https://github.com/microsoft/onnxruntime-genai/pull/1916
• Disable CUDA graph for Phi LongRoPE models with IF nodes on TRT-RTX by @anujj in https://github.com/microsoft/onnxruntime-genai/pull/1921
• Add support for CUDA and CPU arch for Qwen-2.5-VL and Fara-7B by @apsonawane in https://github.com/microsoft/onnxruntime-genai/pull/1919
• Add Gemma-3 vision tutorial to ONNX Runtime GenAI by @kunal-vaishnavi in https://github.com/microsoft/onnxruntime-genai/pull/1793
• Quark GPT-OSS support by @thpereir in https://github.com/microsoft/onnxruntime-genai/pull/1903
• Fix sliding window alignment regression in QNN models by @apsonawane in https://github.com/microsoft/onnxruntime-genai/pull/1938
• AMD RyzenAI EP Support by @akholodnamdcom in https://github.com/microsoft/onnxruntime-genai/pull/1935
• Update README by @natke in https://github.com/microsoft/onnxruntime-genai/pull/1934
• [RyzenAI] Non-pruned models backward compatibility by @akholodnamdcom in https://github.com/microsoft/onnxruntime-genai/pull/1942
• [VitisAI] EP loader by @akholodnamdcom in https://github.com/microsoft/onnxruntime-genai/pull/1918
• Set default top_k and top_p if it is None by @xiaoyu-work in https://github.com/microsoft/onnxruntime-genai/pull/1944
• Ensure dlls are signed in the c and nuget packages. by @baijumeswani in https://github.com/microsoft/onnxruntime-genai/pull/1947
• Bump torch from 2.7.1 to 2.7.1+cpu in /test/python/directml/torch by @dependabot[bot] in https://github.com/microsoft/onnxruntime-genai/pull/1868
• Add linker flags for 16 KB page size on Android by @sheetalarkadam in https://github.com/microsoft/onnxruntime-genai/pull/1860
• Only manually load DLLs if onnxruntime.dll is not already loaded. by @chemwolf6922 in https://github.com/microsoft/onnxruntime-genai/pull/1800
• Add a doc showing how to run GPT OSS 20B with WebGPU by @natke in https://github.com/microsoft/onnxruntime-genai/pull/1945
• Add C#, Java, and Objective-C APIs for Config by @kunal-vaishnavi in https://github.com/microsoft/onnxruntime-genai/pull/1946
• Fix GatherBlockQuantized node to support symmetric quantized LM_HEAD by @sushraja-msft in https://github.com/microsoft/onnxruntime-genai/pull/1951
• Fix QMoE blockwise quantization support for TRT-RTX execution provider by @anujj in https://github.com/microsoft/onnxruntime-genai/pull/1926
• Revert "Add a doc showing how to run GPT OSS 20B with WebGPU" by @kunal-vaishnavi in https://github.com/microsoft/onnxruntime-genai/pull/1950
• Add custom model path support for unit tests by @mpasumarthi-git in https://github.com/microsoft/onnxruntime-genai/pull/1917
• fix: patch llguidance to remove reference to ring crate by @sanaa-hamel-microsoft in https://github.com/microsoft/onnxruntime-genai/pull/1948
• Implement graph models for EPs by @qjia7 in https://github.com/microsoft/onnxruntime-genai/pull/1895
• Update handling EOS token id detection by @kunal-vaishnavi in https://github.com/microsoft/onnxruntime-genai/pull/1925
• Remove onnxruntime-genai-cuda from the foundry package by @baijumeswani in https://github.com/microsoft/onnxruntime-genai/pull/1954
• Include linux builds in the foundry ort-genai package by @baijumeswani in https://github.com/microsoft/onnxruntime-genai/pull/1955
• Support pre-registered plug-in NvTensorRtRtx execution provider library by @anujj in...