[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"extension:v3:en:onnx":3},{"resolvedFromAlias":4,"canonicalExt":5,"ext":5,"name":6,"category":7,"categoryName":8,"updatedAt":9,"popularity":10,"summary":11,"howToOs":12,"quickAnswer":18,"answerIntro":19,"whatIs":20,"description":21,"furtherReading":22,"openInstructions":44,"commonIssues":59,"securityNotes":78,"faq":82,"aliases":98,"mimeTypes":99,"relatedExtensions":100,"breadcrumbs":119,"categoryAnchor":129,"categoryFuturePath":130,"metaDescription":131,"availableHowToOs":132,"openOnDeviceLinks":133,"cannotOpenReasons":149,"cannotOpenFixes":150,"convertOptions":151,"securityAffiliateMessaging":152,"securityAffiliates":153},false,"onnx","ONNX (Open Neural Network Exchange) model","ml","Machine learning","2026-06-12T08:19:11.645Z",45,".onnx is a serialized machine-learning model file in the Open Neural Network Exchange (ONNX) format, used to move models between tools and run inference in different environments. It’s commonly loaded by ONNX Runtime and other ONNX-capable ML software rather than “opened” like a document.",[13,14,15,16,17],"windows","mac","linux","ios","android","To open .ONNX files on Windows, if you want to inspect the model graph (layers, operators, shapes), open the .onnx in Netron or another ONNX-capable graph viewer.","To work with a .onnx file, pick a tool that matches your goal: use Netron (or another ONNX graph viewer) to inspect the model graph; use ONNX Runtime when you want to run inference from a script or notebook; use an ONNX-aware environment like Wolfram Language if you need import/export in that ecosystem. On desktop you usually load the file from inside the tool rather than by double-clicking it.","An ONNX file stores a model as a protobuf-based binary representation of an ONNX computation graph. ONNX tooling commonly infers the format from the .onnx extension when loading/saving models. The file is used for model interchange and deployment across supported frameworks and runtimes.","ONNX (Open Neural Network Exchange) is an open standard format for representing machine-learning models, designed to improve interoperability so a model produced in one tool can be used in another. In practice, a .onnx file is the “serialized model” that you pass to runtimes or tooling that understand ONNX graphs.",[23,26,29,32,35,38,41],{"title":24,"url":25},"Open Neural Network Exchange (Wikipedia)","https://en.wikipedia.org/wiki/Open_Neural_Network_Exchange",{"title":27,"url":28},"Serialization - ONNX 1.22.0 documentation","https://onnx.ai/onnx/api/serialization.html",{"title":30,"url":31},"Load and Run an ONNX Model (onnx-r article)","https://onnx.ai/onnx-r/articles/onnx_model_zoo.html",{"title":33,"url":34},"ONNX Runtime Python API summary","https://onnxruntime.ai/docs/api/python/api_summary",{"title":36,"url":37},"Netron — viewer for ONNX and other models","https://github.com/lutzroeder/netron",{"title":39,"url":40},"ONNX—Wolfram Documentation","https://reference.wolfram.com/language/ref/format/ONNX.html",{"title":42,"url":43},"onnx/models GitHub repository (ONNX Model Zoo)","https://github.com/onnx/models",{"windows":45,"macos":49,"linux":53,"ios":56,"android":58},[46,47,48],"If you want to inspect the model graph (layers, operators, shapes), open the .onnx in Netron or another ONNX-capable graph viewer.","If you want to run the model, use ONNX Runtime: install it in your Python environment and load the .onnx file with the ONNX Runtime Python API to create an inference session.","If you want to import or convert within a computational environment, you can also use Wolfram Language’s ONNX Import/Export support.",[50,51,52],"Open the .onnx in Netron (or a similar viewer) to explore the graph without writing code.","Use ONNX Runtime (Python) to load the .onnx file and run inference via the ONNX Runtime Python API.","Alternatively, use Wolfram Language to import/export ONNX models if that fits your workflow.",[54,51,55],"Use Netron or another ONNX graph viewer to inspect the .onnx structure on a desktop.","For examples of loading and running models and obtaining .onnx models, refer to the ONNX site’s tutorial material (including the Model Zoo workflow).",[57],"There is no guaranteed “open by tapping” support described in the cited sources; for practical use, transfer the .onnx file to a desktop environment and load it with ONNX Runtime or another ONNX-capable tool.",[57],[60,66,72],{"title":61,"description":62,"steps":63},"Double-clicking the file doesn’t open anything useful",".onnx is a serialized ML model (protobuf) meant to be loaded by ML software; most operating systems don’t have a default app to “open” it like a document.",[64,65],"Decide your goal: inspect the graph (use Netron or another ONNX viewer), run inference (use ONNX Runtime), or import/export in a specific toolchain (for example Wolfram Language).","Load the file from within that tool (for example, create an ONNX Runtime inference session in Python and point it to the .onnx path).",{"title":67,"description":68,"steps":69},"The model loads, but you can’t easily “view” it as a human-readable file","ONNX models are protobuf-based binary files, so opening them in a text editor will look like garbled data.",[70,71],"Use ONNX-oriented tooling that understands the model graph rather than a plain text editor.","For a quick visual overview, open the file in Netron; for deeper analysis or execution, use ONNX Runtime or your framework’s ONNX tooling.",{"title":73,"description":74,"steps":75},"You downloaded a model but don’t know if it’s really an ONNX model file","Many distributions refer to a file named model.onnx as the serialized ONNX model; if the extension is missing or changed, tools may not infer it correctly.",[76,77],"Verify the file is intended as an ONNX model (for example, from a source that explicitly describes model.onnx as the serialized ONNX model).","Ensure the filename ends with .onnx so ONNX APIs that infer format from extension treat it as an ONNX model.",[79,80,81],"A .onnx file is a protobuf-serialized model graph; treat untrusted models as untrusted input because parsers and runtimes must deserialize and process complex structured data.","Models from public repositories (for example, model zoos) should be sourced carefully; use only models you trust and validate behavior before using them in production inference pipelines.","Because ONNX files represent computation graphs used for inference, loading an unexpected model can lead to undesired outputs or resource-heavy execution; test in a controlled environment first.",[83,86,89,92,95],{"question":84,"answer":85},"Is an .onnx file a dataset or a trained model?","It is typically a trained model stored as an ONNX computation graph, serialized as a protobuf-based binary file.",{"question":87,"answer":88},"What software can run inference from a .onnx model?","ONNX Runtime is designed to load and run inference on models in ONNX graph format (commonly stored as .onnx files).",{"question":90,"answer":91},"How can I inspect or visualize an .onnx model?","Use an ONNX graph viewer such as Netron to explore operators, tensors, and metadata; it is often the quickest way to “look inside” a model without writing code.",{"question":93,"answer":94},"Can I import/export .onnx models in Wolfram Language?","Yes. Wolfram documentation describes ONNX as a Protobuf-based binary model format and documents importing and exporting .onnx models.",{"question":96,"answer":97},"Does .onnx have an official MIME type?","The provided sources include the IANA Media Types registry as the authoritative place to confirm registered media types; no specific ONNX MIME type is cited in the provided snippets.",[],[],[101,107,113],{"ext":102,"name":103,"category":7,"categoryName":8,"popularity":104,"summary":105,"howToOs":106},"safetensors","SafeTensors tensor checkpoint format",55,".safetensors stores machine-learning tensor weights in a simple binary container with a small JSON header and raw tensor bytes. It is commonly used for sharing and loading model checkpoints (especially in the Hugging Face ecosystem) with an emphasis on safer, validation-friendly loading.",[13,14,15,16,17],{"ext":108,"name":109,"category":7,"categoryName":8,"popularity":110,"summary":111,"howToOs":112},"gguf","GGUF (GGML Universal Format) model file",35,".gguf is a binary file format used to store machine-learning models (tensors plus metadata), commonly for running LLMs with GGUF-capable local tools and runtimes (llama.cpp is a well-known example). It is designed for fast saving/loading and is often used to distribute quantized models for inference.",[13,14,15,16,17],{"ext":114,"name":115,"category":7,"categoryName":8,"popularity":116,"summary":117,"howToOs":118},"ml20","Machine Learning Model Format 2020",15,".ml20 files are used for storing machine learning models, typically created by specific machine learning frameworks.",[13,14,15,16,17],[120,123,126],{"label":121,"to":122},"Home","/",{"label":124,"to":125},"File Extension Index","/file-extension",{"label":127,"to":128},".ONNX","/file-extension/onnx","category-ml","/category/ml","Learn what .ONNX files are, how to open them on every platform, common fixes, and security best practices.",[13,14,15,16,17],[134,137,140,143,146],{"os":13,"label":135,"to":136},"Open .ONNX on Windows","/how-to/open-onnx-on-windows",{"os":14,"label":138,"to":139},"Open .ONNX on Mac","/how-to/open-onnx-on-mac",{"os":15,"label":141,"to":142},"Open .ONNX on Linux","/how-to/open-onnx-on-linux",{"os":16,"label":144,"to":145},"Open .ONNX on iOS","/how-to/open-onnx-on-ios",{"os":17,"label":147,"to":148},"Open .ONNX on Android","/how-to/open-onnx-on-android",[61,67,73],[64,65],[],"untrusted_source",[154,158],{"name":155,"description":156,"affiliateUrl":157},"Avast","Avast offers free and premium antivirus software that protects against viruses, malware, ransomware, and phishing. Scan files before opening them to ensure safety.","https://www.avast.com/lp-aff-consumer-store?expid=inf601",{"name":159,"description":160,"affiliateUrl":161},"Norton","Norton 360 delivers comprehensive antivirus protection, VPN, and identity theft monitoring. Scan files for threats before opening to keep your device secure.","http://buy.norton.com/aff_home?utm_campaign=en-ww_nor_n36_aff_nas_nau_nah_cj_nad_low:_sec_nat_mktc_norton_360"]