Download Latest Version
onnxruntime-linux-x64-gpu_cuda13-1.24.4.tgz (180.3 MB)
Get an email when there's a new version of ONNX Runtime
| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| Parent folder | |||
| onnxruntime-linux-x64-gpu_cuda13-1.24.4.tgz | 2026-03-17 | 180.3 MB | |
| onnxruntime-win-x64-gpu_cuda13-1.24.4.zip | 2026-03-17 | 288.4 MB | |
| onnxruntime-win-x64-gpu-1.24.4.zip | 2026-03-17 | 281.0 MB | |
| onnxruntime-win-x64-1.24.4.zip | 2026-03-17 | 74.4 MB | |
| onnxruntime-win-arm64x-1.24.4.zip | 2026-03-17 | 116.8 MB | |
| onnxruntime-win-arm64-1.24.4.zip | 2026-03-17 | 75.2 MB | |
| onnxruntime-osx-arm64-1.24.4.tgz | 2026-03-17 | 30.9 MB | |
| onnxruntime-linux-x64-gpu-1.24.4.tgz | 2026-03-17 | 205.4 MB | |
| onnxruntime-linux-x64-1.24.4.tgz | 2026-03-17 | 8.2 MB | |
| onnxruntime-linux-aarch64-1.24.4.tgz | 2026-03-17 | 7.2 MB | |
| ONNX Runtime v1.24.4 source code.tar.gz | 2026-03-16 | 283.6 MB | |
| ONNX Runtime v1.24.4 source code.zip | 2026-03-16 | 289.8 MB | |
| README.md | 2026-03-16 | 2.2 kB | |
| Totals: 13 Items | 1.8 GB | 24 | |
This is a patch release for ONNX Runtime 1.24, containing bug fixes and execution provider updates.
Bug Fixes
- Core: Added PCI bus fallback for Linux GPU device discovery in containerized environments (e.g., AKS/Kubernetes) where
nvidia-drmis not loaded but GPU PCI devices are still exposed via sysfs. (#27591) - Plugin EP: Fixed null pointer dereference when iterating output spans in
GetOutputIndex. (#27644) - Plugin EP: Fixed bug that incorrectly assigned duplicate MetaDef IDs to fused nodes in different GraphViews (e.g., then/else branches of an If node), causing session creation to fail with a conflicting kernel error. (#27666)
Execution Provider Updates
- QNN EP: Enabled offline x64 compilation with memhandle IO type by deferring rpcmem library loading to inference time. (#27479)
- QNN EP: Reverted QNN SDK logging verbosity changes that caused segmentation faults on backend destruction. (#27650)
Build and Infrastructure
- Python: Updated
python_requiresfrom>=3.10to>=3.11to reflect dropped Python 3.10 support. (#27354) - Build: Replaced
__builtin_ia32_tpausewith the compiler-portable_tpauseintrinsic to fix cross-compiler portability issues between GCC and LLVM. (#27607)
Full Changelog: v1.24.3...v1.24.4
Contributors
@derdeljan-msft, @adrianlizarraga, @apwojcik, @baijumeswani, @edgchen1, @mocknen, @tianleiwu, @XXXXRT666
