Integrate ONNX 1.22.0 (opset 27) — issue #28752

.agents/skills/ort-build/SKILL.md

@@ -65,6 +65,7 @@ You do **not** need `--update` when only modifying existing `.cc`/`.h` files —
 | `--build_wheel` | Build the Python wheel package |
 | `--use_cuda` | Enable CUDA EP. Requires `--cuda_home`/`--cudnn_home` or `CUDA_HOME`/`CUDNN_HOME` env vars. On Windows, only `cuda_home`/`CUDA_HOME` is validated. |
 | `--target T` | Build a specific CMake target (requires `--build`; e.g., `onnxruntime_common`, `onnxruntime_test_all`) |
+| `--use_webgpu` | Enable WebGPU EP. To run its tests locally on Linux without a GPU, see the `webgpu-local-testing` skill. |
 | `--build_dir` | Build output directory |
 
 ## Build output path

.agents/skills/ort-test/SKILL.md

@@ -79,3 +79,4 @@ Python test naming convention: `test_<method>_<expected_behavior>_[when_<conditi
 - **Redirect test output to a file** (e.g., `> test_output.txt 2>&1`) — output can be large.
 - For C++ tests, verify the build directory exists and a prior build completed before running.
 - Use `--gtest_filter` to run a targeted subset when the full suite takes too long.
+- **Running WebGPU tests locally on Linux without a GPU** — WebGPU op tests build into `onnxruntime_provider_test` and can run against a software Vulkan adapter (Mesa lavapipe). See the `webgpu-local-testing` skill.

.agents/skills/webgpu-local-testing/SKILL.md

@@ -0,0 +1,111 @@
+---
+name: webgpu-local-testing
+description: Build and run ONNX Runtime WebGPU provider tests on Linux WITHOUT a real GPU, using a software Vulkan adapter (Mesa lavapipe). Use when you need to exercise WebGPU EP kernels off-Mac — the Linux webgpu CI leg is build-only, so software Vulkan is how you actually run WebGPU correctness tests locally. SCOPE: lavapipe only validates host-side enforce/shape bugs and MatMul-free kernels; any graph containing MatMul (including the expanded-Attention node tests) crashes lavapipe and runs ONLY on macOS-arm64 Metal, which is the source of truth for those. Covers install (dnf on Azure Linux), the --use_webgpu build flag, the onnxruntime_provider_test target, VK_ICD_FILENAMES, and the lavapipe MatMul crash gotcha.
+---
+
+# Running ONNX Runtime WebGPU Tests Locally on Linux (No GPU)
+
+Reusable knowledge for exercising the **WebGPU execution provider** on a Linux box
+with no physical GPU.
+
+> **Scope**: Linux ORT WebGPU only. macOS uses the Metal backend and a real GPU;
+> this skill is the off-Mac story for running WebGPU EP kernels in CI-less dev loops.
+
+## 1. Why software Vulkan is enough
+
+On Linux, ORT's WebGPU EP runs on **Dawn** with the **Vulkan** backend. Vulkan does
+not require a hardware GPU — **Mesa lavapipe** is a software (CPU) Vulkan adapter that
+Dawn enumerates like any other device. For **EP correctness tests** this is sufficient
+because:
+
+- Many host-side validation paths (shape/broadcast checks, `ORT_ENFORCE`s) fire
+  **before** any shader is dispatched. E.g. the WebGPU broadcast `ORT_ENFORCE` runs
+  host-side, so the failure is observable on a software adapter without ever touching
+  the GPU.
+- Element-wise and broadcasting kernels that do dispatch run correctly (if slowly) on
+  lavapipe, so their numeric output can be validated against the CPU reference.
+
+You are trading speed for not needing hardware. It is **not** a substitute for a real
+GPU on perf-sensitive or driver-specific paths — but for kernel correctness it is the
+practical local loop.
+
+> **Scope — what lavapipe can and cannot validate.** Software Vulkan covers (a)
+> **host-side** failures (shape/broadcast `ORT_ENFORCE`s that fire *before* any shader
+> dispatch) and (b) **MatMul-free** kernels that dispatch. It does **NOT** cover any
+> graph that contains a **MatMul** — the MatMul family crashes lavapipe's LLVM JIT (see
+> §5). This explicitly includes the motivating expanded-Attention node tests
+> (`test_attention_4d_softcap_neginf_mask_expanded`): they decompose to
+> `softmax(Q·Kᵀ + bias)·V`, which **contains MatMuls**, so they **cannot run on
+> lavapipe**. For those, **macOS-arm64 Metal is the source of truth**. Concretely, the
+> #28969 WebGPU broadcast-underflow fix was validated on lavapipe via a standalone
+> Add-broadcast `OpTester` proxy (a host-side enforce/shape path) — **NOT** via the
+> expanded-Attention node test. **Never** run lavapipe green and conclude an
+> Attention/MatMul fix is validated off-Mac.
+
+## 2. Install the software Vulkan stack (Azure Linux — use `dnf`, NOT `apt`)
+
+```bash
+dnf install -y mesa-vulkan-drivers vulkan-loader
+# optional, for sanity-checking the adapter:
+dnf install -y vulkan-tools && vulkaninfo | head
+```
+
+`mesa-vulkan-drivers` provides lavapipe; `vulkan-loader` provides the ICD loader.
+The lavapipe ICD manifest lands at `/usr/share/vulkan/icd.d/lvp_icd.<arch>.json` —
+`lvp_icd.x86_64.json` on x86_64, `lvp_icd.aarch64.json` on arm64. The examples below
+use the x86_64 name; substitute your arch, or glob it:
+`VK_ICD_FILENAMES=$(echo /usr/share/vulkan/icd.d/lvp_icd.*.json)`.
+
+## 3. Build with `--use_webgpu`
+
+```bash
+./build.sh --config Release --parallel --use_webgpu
+```
+
+See the `ort-build` skill for general build phases and flags.
+
+## 4. Run the WebGPU provider tests
+
+WebGPU operator/kernel tests are **provider op tests** — they build into the
+`onnxruntime_provider_test` target (**NOT** `onnxruntime_test_all`; see the `ort-test`
+skill for the executable taxonomy). Point the Vulkan loader at the lavapipe ICD and
+select a subset with `--gtest_filter`:
+
+```bash
+cd build/Linux/Release
+VK_ICD_FILENAMES=/usr/share/vulkan/icd.d/lvp_icd.x86_64.json \
+  ./onnxruntime_provider_test --gtest_filter="*WebGPU*"
+```
+
+`VK_ICD_FILENAMES` forces Vulkan to load **only** lavapipe, so the run is
+deterministic regardless of what else is installed.
+
+## 5. Gotcha: lavapipe crashes on the MatMul family
+
+`MathOpTest.MatMulFloatType` (and other MatMul-family tests) crash lavapipe with:
+
+```
+LLVM ERROR: Instruction Combining did not reach a fixpoint after 1 iterations
+```
+
+This is a **pre-existing limitation of software Vulkan (Mesa lavapipe's LLVM JIT)**,
+not an ORT bug. **Exclude the MatMul family** from broad lavapipe runs:
+
+```bash
+VK_ICD_FILENAMES=/usr/share/vulkan/icd.d/lvp_icd.x86_64.json \
+  ./onnxruntime_provider_test --gtest_filter="*WebGPU*:-*MatMul*"
+```
+
+## 6. Why this matters: the Linux webgpu CI leg is build-only
+
+The Linux webgpu CI leg (`py-linux-webgpu-stage.yml`) **only builds** — it does not run
+WebGPU kernels. A green Linux webgpu leg therefore does **not** mean any WebGPU test
+actually executed. The macOS-arm64 webgpu leg is the only CI leg that runs WebGPU
+backend node tests. So a local lavapipe run is the practical way to **actually exercise
+WebGPU kernels off-Mac** before you push.
+
+But mind the §1 scope: lavapipe covers host-side enforce/shape paths and **MatMul-free**
+kernels only. Any **MatMul-containing** graph — including the expanded-Attention node
+tests (`test_attention_4d_softcap_neginf_mask_expanded`) — crashes lavapipe and runs
+**only** on the macOS-arm64 Metal leg, which is the source of truth for those. A green
+lavapipe run never validates a MatMul/Attention fix off-Mac.

cmake/deps.txt

@@ -34,7 +34,10 @@ microsoft_gsl;https://github.com/microsoft/GSL/archive/refs/tags/v4.2.1.zip;1094
 microsoft_wil;https://github.com/microsoft/wil/archive/refs/tags/v1.0.250325.1.zip;826c8bd47c2258ec61b8b218e031e5b33d27f761
 mimalloc;https://github.com/microsoft/mimalloc/archive/refs/tags/v2.1.1.zip;d5ee7d34223d0567892db5179849939c8769dc41
 mp11;https://github.com/boostorg/mp11/archive/refs/tags/boost-1.82.0.zip;9bc9e01dffb64d9e0773b2e44d2f22c51aace063
-onnx;https://github.com/onnx/onnx/archive/refs/tags/v1.21.0.zip;321d4acc807c8e0fb0bbcc0424a143dffde1e846
+# ONNX pinned to the 1.22.0 final release (tag v1.22.0). rc2->final changed only CI
+# workflow yml + VERSION_NUMBER (no operator-schema, opset, or testdata change), so it is
+# pure version plumbing. The 7 CI requirements.txt files are pinned to onnx==1.22.0 (now on PyPI).
+onnx;https://github.com/onnx/onnx/archive/refs/tags/v1.22.0.zip;2b2cd58ac7a26df5371266149e0c76776330cdf1
 # Use the latest commit of 10.9-GA
 onnx_tensorrt;https://github.com/onnx/onnx-tensorrt/archive/d5dce67db7c2e64b07e055571f5ec06f7f254de2.zip;01114d3b67650857281fa50faa2e412130a63b69
 protobuf;https://github.com/protocolbuffers/protobuf/archive/refs/tags/v21.12.zip;7cf2733949036c7d52fda017badcab093fe73bfa

cmake/patches/onnx/onnx.patch

@@ -1,80 +1,44 @@
 diff --git a/CMakeLists.txt b/CMakeLists.txt
-index 044996e..ded7e39 100644
+index 791e24b..5f861a1 100644
 --- a/CMakeLists.txt
 +++ b/CMakeLists.txt
-@@ -53,6 +53,7 @@ option(ONNX_USE_LITE_PROTO "Use lite protobuf instead of full." OFF)
- option(ONNX_DISABLE_EXCEPTIONS "Disable exception handling." OFF)
+@@ -46,6 +46,7 @@ option(ONNX_DISABLE_EXCEPTIONS "Disable exception handling." OFF)
  option(ONNX_DISABLE_STATIC_REGISTRATION "Disable static registration for ONNX operator schemas." OFF)
  option(ONNX_USE_UNITY_BUILD "Enable Unity (Jumbo) build for" OFF)
+ option(ONNX_PYODIDE_BUILD "Build ONNX for Pyodide/Emscripten (wasm32)." OFF)
 +option(ONNX_MINIMAL_BUILD "Build only essential ONNX components" OFF)
  option(ONNX_INSTALL "Install ONNX targets, headers, and CMake config files" ON)
  if(WIN32)
-   option(ONNX_USE_MSVC_STATIC_RUNTIME "Build with MSVC static runtime" OFF)
-@@ -399,14 +400,28 @@ relative_protobuf_generate_cpp(ONNX_PROTO_SRCS
-                                onnx/onnx-operators.in.proto
-                                onnx/onnx-data.in.proto)
-
--file(GLOB_RECURSE __tmp_srcs "${ONNX_ROOT}/onnx/*.h" "${ONNX_ROOT}/onnx/*.cc")
--file(GLOB_RECURSE onnx_gtests_src "${ONNX_ROOT}/onnx/test/cpp/*.h"
--    "${ONNX_ROOT}/onnx/test/cpp/*.cc"
--    "${ONNX_ROOT}/onnx/backend/test/cpp/*.cc"
--    "${ONNX_ROOT}/onnx/backend/test/cpp/*.h")
--list(REMOVE_ITEM __tmp_srcs "${ONNX_ROOT}/onnx/cpp2py_export.cc")
--list(REMOVE_ITEM __tmp_srcs ${onnx_gtests_src} "${ONNX_ROOT}/onnx/test/cmake/main.cc")
--list(APPEND ONNX_SRCS ${__tmp_srcs})
+   option(ONNX_USE_MSVC_STATIC_RUNTIME "Build with MSVC static runtime. Ignored if CMAKE_MSVC_RUNTIME_LIBRARY is defined." OFF)
+@@ -515,7 +516,22 @@ add_dependencies(onnx_proto onnx_proto_gen)
+ # Core library. Hand-written sources come from target_sources(onnx ...) in the
+ # subdirectories; the protobuf objects are pulled in from onnx_proto.
+ add_library(onnx $<TARGET_OBJECTS:onnx_proto>)
+-add_subdirectory(onnx)
 +if(ONNX_MINIMAL_BUILD)
 +    message(STATUS "Configuring ONNX minimal build")
-+    set(ONNX_SRCS
-+      "${ONNX_ROOT}/onnx/common/common.h"
-+      "${ONNX_ROOT}/onnx/defs/data_type_utils.h"
-+      "${ONNX_ROOT}/onnx/defs/data_type_utils.cc"
++    # ORT's minimal/WebAssembly builds only need ONNX's type utilities and the
++    # generated protobuf sources (linked via onnx_proto above); skip pulling in
++    # the full operator/schema sources via add_subdirectory(onnx).
++    target_sources(onnx PRIVATE
++        "${ONNX_ROOT}/onnx/defs/data_type_utils.cc"
 +    )
 +    # Ensure ONNX_ML is treated as ON for minimal build consistency with ORT's file
 +    set(ONNX_ML ON CACHE BOOL "Enable traditional ML API." FORCE)
 +    # Minimal build doesn't need Python or tests
 +    set(ONNX_BUILD_PYTHON OFF CACHE BOOL "Build Python binaries" FORCE)
 +    set(ONNX_BUILD_TESTS OFF CACHE BOOL "Build ONNX C++ APIs Tests" FORCE)
 +else()
-+    file(GLOB_RECURSE __tmp_srcs "${ONNX_ROOT}/onnx/*.h" "${ONNX_ROOT}/onnx/*.cc")
-+    file(GLOB_RECURSE onnx_gtests_src "${ONNX_ROOT}/onnx/test/cpp/*.h"
-+        "${ONNX_ROOT}/onnx/test/cpp/*.cc"
-+        "${ONNX_ROOT}/onnx/backend/test/cpp/*.cc"
-+        "${ONNX_ROOT}/onnx/backend/test/cpp/*.h")
-+    list(REMOVE_ITEM __tmp_srcs "${ONNX_ROOT}/onnx/cpp2py_export.cc")
-+    list(REMOVE_ITEM __tmp_srcs ${onnx_gtests_src} "${ONNX_ROOT}/onnx/test/cmake/main.cc")
-+    list(APPEND ONNX_SRCS ${__tmp_srcs})
++    add_subdirectory(onnx)
 +endif()
-
- set(LINKED_PROTOBUF_TARGET protobuf::libprotobuf)
- if(ONNX_USE_LITE_PROTO)
-diff --git a/cmake/Utils.cmake b/cmake/Utils.cmake
-index 1987edd..04b3088 100644
---- a/cmake/Utils.cmake
-+++ b/cmake/Utils.cmake
-@@ -103,18 +103,7 @@ endfunction()
-
- function(add_onnx_compile_options target)
-   if(MSVC)
--    # For disabling Protobuf related warnings
--    set(protobuf_warnings
--        /wd4146 # unary minus operator applied to unsigned type, result still
--                # unsigned
--        /wd4244 # 'argument': conversion from 'google::protobuf::uint64' to
--                # 'int', possible loss of data
--        /wd4267 # Conversion from 'size_t' to 'int', possible loss of data
--        /wd4141 # 'inline': used more than once
--        /wd4047 # '=': 'uintptr_t' differs in levels of indirection from 'void *'
--    )
-     add_msvc_runtime_flag(${target})
--    target_compile_options(${target} PUBLIC ${protobuf_warnings})
-     if(ONNX_WERROR)
-       target_compile_options(${target} PRIVATE "/WX")
-     endif()
+ add_dependencies(onnx onnx_proto)
+ set_target_properties(onnx PROPERTIES CXX_VISIBILITY_PRESET hidden)
+ add_onnx_global_defines(onnx)
 diff --git a/onnx/defs/nn/old.cc b/onnx/defs/nn/old.cc
-index a6a8a83..153da87 100644
+index 3887169..8432613 100644
 --- a/onnx/defs/nn/old.cc
 +++ b/onnx/defs/nn/old.cc
-@@ -4026,7 +4026,6 @@ ONNX_OPERATOR_SET_SCHEMA(
+@@ -4078,7 +4078,6 @@ ONNX_OPERATOR_SET_SCHEMA(
      GroupNormalization,
      18,
      OpSchema()

cmake/vcpkg-ports/onnx/binskim.patch

@@ -1,80 +1,44 @@
 diff --git a/CMakeLists.txt b/CMakeLists.txt
-index 044996e..ded7e39 100644
+index 791e24b..5f861a1 100644
 --- a/CMakeLists.txt
 +++ b/CMakeLists.txt
-@@ -53,6 +53,7 @@ option(ONNX_USE_LITE_PROTO "Use lite protobuf instead of full." OFF)
- option(ONNX_DISABLE_EXCEPTIONS "Disable exception handling." OFF)
+@@ -46,6 +46,7 @@ option(ONNX_DISABLE_EXCEPTIONS "Disable exception handling." OFF)
  option(ONNX_DISABLE_STATIC_REGISTRATION "Disable static registration for ONNX operator schemas." OFF)
  option(ONNX_USE_UNITY_BUILD "Enable Unity (Jumbo) build for" OFF)
+ option(ONNX_PYODIDE_BUILD "Build ONNX for Pyodide/Emscripten (wasm32)." OFF)
 +option(ONNX_MINIMAL_BUILD "Build only essential ONNX components" OFF)
  option(ONNX_INSTALL "Install ONNX targets, headers, and CMake config files" ON)
  if(WIN32)
-   option(ONNX_USE_MSVC_STATIC_RUNTIME "Build with MSVC static runtime" OFF)
-@@ -399,14 +400,28 @@ relative_protobuf_generate_cpp(ONNX_PROTO_SRCS
-                                onnx/onnx-operators.in.proto
-                                onnx/onnx-data.in.proto)
-
--file(GLOB_RECURSE __tmp_srcs "${ONNX_ROOT}/onnx/*.h" "${ONNX_ROOT}/onnx/*.cc")
--file(GLOB_RECURSE onnx_gtests_src "${ONNX_ROOT}/onnx/test/cpp/*.h"
--    "${ONNX_ROOT}/onnx/test/cpp/*.cc"
--    "${ONNX_ROOT}/onnx/backend/test/cpp/*.cc"
--    "${ONNX_ROOT}/onnx/backend/test/cpp/*.h")
--list(REMOVE_ITEM __tmp_srcs "${ONNX_ROOT}/onnx/cpp2py_export.cc")
--list(REMOVE_ITEM __tmp_srcs ${onnx_gtests_src} "${ONNX_ROOT}/onnx/test/cmake/main.cc")
--list(APPEND ONNX_SRCS ${__tmp_srcs})
+   option(ONNX_USE_MSVC_STATIC_RUNTIME "Build with MSVC static runtime. Ignored if CMAKE_MSVC_RUNTIME_LIBRARY is defined." OFF)
+@@ -515,7 +516,22 @@ add_dependencies(onnx_proto onnx_proto_gen)
+ # Core library. Hand-written sources come from target_sources(onnx ...) in the
+ # subdirectories; the protobuf objects are pulled in from onnx_proto.
+ add_library(onnx $<TARGET_OBJECTS:onnx_proto>)
+-add_subdirectory(onnx)
 +if(ONNX_MINIMAL_BUILD)
 +    message(STATUS "Configuring ONNX minimal build")
-+    set(ONNX_SRCS
-+      "${ONNX_ROOT}/onnx/common/common.h"
-+      "${ONNX_ROOT}/onnx/defs/data_type_utils.h"
-+      "${ONNX_ROOT}/onnx/defs/data_type_utils.cc"
++    # ORT's minimal/WebAssembly builds only need ONNX's type utilities and the
++    # generated protobuf sources (linked via onnx_proto above); skip pulling in
++    # the full operator/schema sources via add_subdirectory(onnx).
++    target_sources(onnx PRIVATE
++        "${ONNX_ROOT}/onnx/defs/data_type_utils.cc"
 +    )
 +    # Ensure ONNX_ML is treated as ON for minimal build consistency with ORT's file
 +    set(ONNX_ML ON CACHE BOOL "Enable traditional ML API." FORCE)
 +    # Minimal build doesn't need Python or tests
 +    set(ONNX_BUILD_PYTHON OFF CACHE BOOL "Build Python binaries" FORCE)
 +    set(ONNX_BUILD_TESTS OFF CACHE BOOL "Build ONNX C++ APIs Tests" FORCE)
 +else()
-+    file(GLOB_RECURSE __tmp_srcs "${ONNX_ROOT}/onnx/*.h" "${ONNX_ROOT}/onnx/*.cc")
-+    file(GLOB_RECURSE onnx_gtests_src "${ONNX_ROOT}/onnx/test/cpp/*.h"
-+        "${ONNX_ROOT}/onnx/test/cpp/*.cc"
-+        "${ONNX_ROOT}/onnx/backend/test/cpp/*.cc"
-+        "${ONNX_ROOT}/onnx/backend/test/cpp/*.h")
-+    list(REMOVE_ITEM __tmp_srcs "${ONNX_ROOT}/onnx/cpp2py_export.cc")
-+    list(REMOVE_ITEM __tmp_srcs ${onnx_gtests_src} "${ONNX_ROOT}/onnx/test/cmake/main.cc")
-+    list(APPEND ONNX_SRCS ${__tmp_srcs})
++    add_subdirectory(onnx)
 +endif()
-
- set(LINKED_PROTOBUF_TARGET protobuf::libprotobuf)
- if(ONNX_USE_LITE_PROTO)
-diff --git a/cmake/Utils.cmake b/cmake/Utils.cmake
-index 1987edd..04b3088 100644
---- a/cmake/Utils.cmake
-+++ b/cmake/Utils.cmake
-@@ -103,18 +103,7 @@ endfunction()
-
- function(add_onnx_compile_options target)
-   if(MSVC)
--    # For disabling Protobuf related warnings
--    set(protobuf_warnings
--        /wd4146 # unary minus operator applied to unsigned type, result still
--                # unsigned
--        /wd4244 # 'argument': conversion from 'google::protobuf::uint64' to
--                # 'int', possible loss of data
--        /wd4267 # Conversion from 'size_t' to 'int', possible loss of data
--        /wd4141 # 'inline': used more than once
--        /wd4047 # '=': 'uintptr_t' differs in levels of indirection from 'void *'
--    )
-     add_msvc_runtime_flag(${target})
--    target_compile_options(${target} PUBLIC ${protobuf_warnings})
-     if(ONNX_WERROR)
-       target_compile_options(${target} PRIVATE "/WX")
-     endif()
+ add_dependencies(onnx onnx_proto)
+ set_target_properties(onnx PROPERTIES CXX_VISIBILITY_PRESET hidden)
+ add_onnx_global_defines(onnx)
 diff --git a/onnx/defs/nn/old.cc b/onnx/defs/nn/old.cc
-index a6a8a83..153da87 100644
+index 3887169..8432613 100644
 --- a/onnx/defs/nn/old.cc
 +++ b/onnx/defs/nn/old.cc
-@@ -4026,7 +4026,6 @@ ONNX_OPERATOR_SET_SCHEMA(
+@@ -4078,7 +4078,6 @@ ONNX_OPERATOR_SET_SCHEMA(
      GroupNormalization,
      18,
      OpSchema()

cmake/vcpkg-ports/onnx/portfile.cmake

@@ -3,8 +3,8 @@ vcpkg_check_linkage(ONLY_STATIC_LIBRARY)
 vcpkg_from_github(
     OUT_SOURCE_PATH SOURCE_PATH
     REPO onnx/onnx
-    REF "v${VERSION}"
-    SHA512 3cee4c0fbc9e260e360a62a59e324e0b127a5749f958e0704989b407a4c1179c637ef86e41a406e7868537a62a11a821e3433005eb0725f979145f8d514926bd
+    REF "v1.22.0"
+    SHA512 13fafff073a8e0bcf67fd06195da979c3c6273b3f5fdd25f5e5c39ad6af20eefe107f13dc73d52b6021e986d87401e46121e00bba8fbd09f098334e34f78462f
     PATCHES
         fix-cmakelists.patch
         fix-dependency-protobuf.patch

cmake/vcpkg-ports/onnx/vcpkg.json

@@ -1,6 +1,6 @@
 {
   "name": "onnx",
-  "version-semver": "1.21.0",
+  "version-semver": "1.22.0",
   "port-version": 0,
   "description": "Open standard for machine learning interoperability",
   "homepage": "https://onnx.ai",