{"full_name":"microsoft/onnxruntime","owner":"microsoft","name":"onnxruntime","shard":"1","url":"https://github.com/microsoft/onnxruntime","description":"ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator","homepage":"https://onnxruntime.ai","topics":["deep-learning","onnx","neural-networks","machine-learning","ai-framework","hardware-acceleration","pytorch","tensorflow","scikit-learn"],"tag_pages":["deep-learning","hardware-acceleration","machine-learning","neural-networks","onnx","pytorch","scikit-learn","tensorflow"],"language":"C++","license":"MIT","avatar":"https://avatars.githubusercontent.com/u/6154722?v=4","created_at":"2018-11-10T02:22:53Z","pushed_at":"2026-11-15T00:32:24Z","metrics":{"stars":22317,"forks":4449,"watchers":264,"issues":8895,"prs":21801,"releases":92,"issues_open":937,"issues_closed":7958,"prs_open":1034,"prs_closed":3565,"prs_merged":17202,"discussions":932},"growth":{"day":{"stars":11,"forks":0,"base_day":"2026-11-15"},"week":{"stars":99,"forks":48,"base_day":"2026-11-04"},"month":{"stars":262,"forks":148,"base_day":"2026-10-17"}},"history":{"dates":["2026-09-06","2026-09-07","2026-09-08","2026-09-09","2026-09-10","2026-09-11","2026-09-12","2026-09-13","2026-09-14","2026-09-15","2026-09-16","2026-09-17","2026-09-18","2026-09-19","2026-09-20","2026-09-21","2026-09-22","2026-09-23","2026-09-24","2026-09-25","2026-09-26","2026-09-27","2026-09-28","2026-09-29","2026-09-30","2026-10-01","2026-10-02","2026-10-03","2026-10-04","2026-10-05","2026-10-06","2026-10-07","2026-10-08","2026-10-09","2026-10-10","2026-10-11","2026-10-12","2026-10-13","2026-10-14","2026-10-15","2026-10-16","2026-10-17","2026-10-18","2026-10-19","2026-10-20","2026-10-21","2026-10-22","2026-10-23","2026-10-24","2026-10-25","2026-10-26","2026-10-27","2026-10-28","2026-10-29","2026-10-30","2026-10-31","2026-11-01","2026-11-02","2026-11-03","2026-11-04","2026-11-10","2026-11-11","2026-11-12","2026-11-13","2026-11-14","2026-11-15","2026-11-16"],"stars":[21623,21638,21652,21656,21660,21664,21675,21707,21726,21755,21763,21767,21790,21797,21804,21814,21826,21833,21849,21856,21865,21876,21887,21894,21903,21907,21917,21927,21935,21932,21943,21948,21977,21981,21998,22022,22025,22032,22038,22041,22050,22055,22072,22080,22083,22099,22107,22118,22121,22131,22145,22152,22162,22175,22181,22184,22187,22195,22211,22218,22273,22285,22293,22296,22300,22306,22317],"forks":[4168,4175,4176,4177,4180,4182,4186,4194,4198,4201,4205,4208,4208,4210,4210,4214,4219,4221,4229,4230,4232,4233,4233,4233,4241,4242,4248,4251,4253,4253,4260,4264,4270,4269,4274,4280,4281,4283,4285,4289,4291,4301,4307,4311,4315,4323,4328,4336,4344,4346,4354,4359,4361,4372,4375,4378,4382,4387,4397,4401,4432,4438,4443,4445,4445,4449,4449],"watchers":[262,262,262,262,262,262,262,263,263,263,263,263,264,264,264,264,264,264,264,264,264,264,264,264,264,264,264,264,264,264,264,264,264,264,264,264,264,264,265,265,265,265,265,265,265,265,265,265,265,265,265,266,264,263,263,263,263,263,264,264,null,264,264,264,264,264,264],"issues_open":[null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,925,932,931,933,934,937],"prs_open":[null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,988,1005,1007,1020,1032,1034]},"boards":{},"same_owner":[{"full_name":"microsoft/markitdown","name":"markitdown","shard":"1","boards":{"daily":151,"weekly":176,"monthly":103,"total":36}},{"full_name":"microsoft/vscode","name":"vscode","shard":"1","boards":{"daily_watchers":21,"total":37}},{"full_name":"microsoft/PowerToys","name":"PowerToys","shard":"1","boards":{"daily_watchers":111,"total":81}},{"full_name":"microsoft/generative-ai-for-beginners","name":"generative-ai-for-beginners","shard":"1","boards":{"daily_forks":40,"total":98}},{"full_name":"microsoft/TypeScript","name":"TypeScript","shard":"1","boards":{"daily_watchers":120,"total":120}},{"full_name":"microsoft/terminal","name":"terminal","shard":"1","boards":{"total":133}},{"full_name":"microsoft/playwright","name":"playwright","shard":"1","boards":{"total":143}},{"full_name":"microsoft/Web-Dev-For-Beginners","name":"Web-Dev-For-Beginners","shard":"1","boards":{"total":155}},{"full_name":"microsoft/ML-For-Beginners","name":"ML-For-Beginners","shard":"1","boards":{"total":173}},{"full_name":"microsoft/AI-For-Beginners","name":"AI-For-Beginners","shard":"1","boards":{}},{"full_name":"microsoft/AirSim","name":"AirSim","shard":"1","boards":{}},{"full_name":"microsoft/BitNet","name":"BitNet","shard":"1","boards":{}},{"full_name":"microsoft/Bringing-Old-Photos-Back-to-Life","name":"Bringing-Old-Photos-Back-to-Life","shard":"1","boards":{}},{"full_name":"microsoft/CNTK","name":"CNTK","shard":"1","boards":{}},{"full_name":"microsoft/Data-Science-For-Beginners","name":"Data-Science-For-Beginners","shard":"1","boards":{}},{"full_name":"microsoft/FluidFramework","name":"FluidFramework","shard":"1","boards":{}},{"full_name":"microsoft/IoT-For-Beginners","name":"IoT-For-Beginners","shard":"1","boards":{}},{"full_name":"microsoft/MMdnn","name":"MMdnn","shard":"1","boards":{}},{"full_name":"microsoft/Ontology-Playground","name":"Ontology-Playground","shard":"1","boards":{}},{"full_name":"microsoft/Procexp-Mac","name":"Procexp-Mac","shard":"1","boards":{"daily_rate":96,"weekly_rate":32,"monthly_rate":87}}],"issues":[{"number":20876,"title":"[Web] WebGPU and WASM Backends Unavailable within Service Worker","comments":67,"created_at":"2024-05-30T23:09:14Z"},{"number":16121,"title":"EfficientNMS_TRT missing attribute class_agnostic w/ TensorRT 8.6","comments":35,"created_at":"2023-05-26T19:24:14Z"},{"number":28284,"title":"Cara - Cara Membatalkan Pinjaman Akulaku","comments":31,"created_at":"2026-04-30T04:36:07Z"},{"number":19846,"title":"C# I need to run the program on NPU （OnnxRuntime + DirectML + NPU），but it failed","comments":30,"created_at":"2024-03-11T09:35:29Z"},{"number":14997,"title":"[Feature Request] 4bit and 2bit and 1bit quantization support","comments":30,"created_at":"2023-03-10T07:37:02Z"},{"number":9313,"title":"ONNXRuntime CPU - Memory spiking continuously (Memory leak)","comments":27,"created_at":"2021-10-08T18:55:05Z"},{"number":886,"title":"onnxruntime slower than xgboost & lightgbm in batch predictions","comments":27,"created_at":"2019-04-23T11:38:39Z"},{"number":21917,"title":"[Feature Request] Add vulkan execution provider","comments":26,"created_at":"2024-08-29T20:59:04Z"}],"discussions":[{"number":34225,"title":"How do I install the CUDA thingie?","comments":10,"category":"EP Q&A","created_at":"2026-11-14T10:06:55Z"},{"number":34187,"title":"MIGraphX EP","comments":2,"category":"EP Q&A","created_at":"2026-11-13T10:23:46Z"},{"number":34073,"title":"Orange Pi Zero 3W (12 GB) – Complete power-off during ONNX Runtime / Piper TTS inference","comments":1,"category":"General","created_at":"2026-11-10T06:08:15Z"},{"number":33878,"title":"YOLOVAL: local game-image checks with ONNX Runtime CPU (paid model showcase)","comments":0,"category":"Show & Tell","created_at":"2026-11-04T17:03:10Z"},{"number":33737,"title":"Proposal: a browser model diagnostics workbench with exportable reproductions","comments":1,"category":"Ideas / Feature Requests","created_at":"2026-10-29T15:11:22Z"},{"number":33665,"title":"Release-bound dependency resolution and native provenance for onnxruntime-win-x64 1.29.0 CPU package","comments":1,"category":"Release Q&A","created_at":"2026-10-28T06:30:18Z"},{"number":33637,"title":"edge-nix: Packaging ONNX Runtime for edge devices using Nix (Arduino UNO Q)","comments":0,"category":"Show & Tell","created_at":"2026-10-27T06:49:28Z"},{"number":33488,"title":"onnxruntime-node@1.21.0 Windows x64：已发布原生文件对应的第三方告知与来源材料","comments":1,"category":"Release Q&A","created_at":"2026-10-20T11:30:11Z"}],"latest_release":{"name":"ONNX Runtime v1.32.0","tag":"v1.32.0","published_at":"2026-11-13T09:36:22Z","body":"ONNX Runtime 1.32.0 reintegrates ONNX 1.23.2 with opset 28 support, introduces CPU/GPU hybrid MoE offloading and shared-layout CPU QMoE kernels, expands device-memory, weightless-compilation, and Python model-package APIs, dramatically accelerates session initialization on large graphs, and strengthens runtime and optimizer reliability. These notes cover changes since ONNX Runtime 1.31.1. CUDA and WebGPU kernel updates are summarized here; detailed provider notes are covered by their separate plugin EP releases.\r\n\r\n## Highlights\r\n\r\n- Reintegrated **ONNX 1.23.2** with **opset 28** operator support (`SpaceToDepth`, `Mod`, signed `BitShift`, and `BFloat16` extensions) and updated **DLPack to v1.3** across all implemented ORT numeric and boolean tensor types ([#33134](https://github.com/microsoft/onnxruntime/pull/33134), [#32362](https://github.com/microsoft/onnxruntime/pull/32362), [#33848](https://github.com/microsoft/onnxruntime/pull/33848)).\r\n- Added a session-scoped **CPU/GPU hybrid MoE offload** planner and execution backend for quantized `QMoE` (NVFP4, INT4, INT3, and INT2) and block-packed `MoEPacked` models, backed by shared-layout AVX-512/AVX2 CPU `QMoE` kernels and memory-mapped expert storage ([#33450](https://github.com/microsoft/onnxruntime/pull/33450), [#33451](https://github.com/microsoft/onnxruntime/pull/33451), [#33452](https://github.com/microsoft/onnxruntime/pull/33452), [#33973](https://github.com/microsoft/onnxruntime/pull/33973)).\r\n- Expanded core APIs with `OrtApi::EpDevice_GetMemoryCapacity`, explicit multi-mode weightless compilation (`ModelCompilationOptions_SetWeightlessMode` and `ep.enable_weightless_mode`), Python bindings for the Model Package APIs, and zero-copy CPU buffer borrowing in `AddExternalInitializers` ([#33447](https://github.com/microsoft/onnxruntime/pull/33447), [#32899](https://github.com/microsoft/onnxruntime/pull/32899), [#33640](https://github.com/microsoft/onnxruntime/pull/33640), [#33877](https://github.com/microsoft/onnxruntime/pull/33877)).\r\n- Eliminated superlinear bottlenecks in allocation planning, constant sharing, graph I/O mapping, and topological sorting—reducing session initialization on 100k-node constant-heavy graphs from minutes to seconds—and added pipelined pinned-memory staging for external GPU weight loading ([#33407](https://github.com/microsoft/onnxruntime/pull/33407), [#33939](https://github.com/microsoft/onnxruntime/pull/33939), [#34009](https://github.com/microsoft/onnxruntime/pull/34009), [#33839](https://github.com/microsoft/onnxruntime/pull/33839)).\r\n- Expanded shared generative AI and quantized contrib operators, including Kimi Delta Attention (`GatedDeltaNet` per-key-dimension decay), right-padded chunked prefill in `GroupQueryAttention`, native GGUF/block-packed operators (`GatherPacked`, `MatMulPacked`, `MoEPacked`), and sparse-attention index sharing ([#33817](https://github.com/microsoft/onnxruntime/pull/33817), [#33503](https://github.com/microsoft/onnxruntime/pull/33503), [#33953](https://github.com/microsoft/onnxruntime/pull/33953), [#33224](https://github.com/microsoft/onnxruntime/pull/33224)).\r\n\r\n## Announcements & Compatibility\r\n\r\n- **CUDA Plugin EP v0.4.0** and **WebGPU Plugin EP v0.6.0** are matched with ONNX Runtime 1.32.0 ([#33214](https://github.com/microsoft/onnxruntime/pull/33214)). For **contrib-op compatibility**, core and the plugin must use the same contributed-operator schemas; building both from the **same ONNX Runtime revision** is recommended. Schema mismatches can cause incorrect execution or crashes and may not be detected during plugin registration.\r\n- **ACL EP removed:** Following its deprecation in 1.31.0, the Arm Compute Library (`ACLExecutionProvider`) execution provider, its build options, and its standalone C registration API have been removed. Python's `get_all_providers()` no longer lists `ACLExecutionProvider`; applications should remove it from explicit provider lists ([#33851](https://github.com/microsoft/onnxruntime","is_prerelease":false},"readme":true,"readme_zh":false,"news":[],"stale":false,"snapshot_at":"2026-11-16T09:23:14Z","history_start":"2026-09-06"}