# ONNX Runtime v1.29.0 — ONNX Runtime v1.29.0

- Product: ONNX Runtime (https://whatsnew.fyi/product/onnx-runtime)
- Vendor: Microsoft
- Date: 2026-08-12
- Version: v1.29.0
- Original notes: https://github.com/microsoft/onnxruntime/releases/tag/v1.29.0
- Permalink: https://whatsnew.fyi/product/onnx-runtime/releases/v1.29.0

What's New is an index, not a publisher: every entry below links to the vendor's own release notes, which are the authoritative source. Entries are labelled where they are hand-curated sample data, pre-releases, or drawn from a secondary source such as a developer blog.

Reuse: the summaries, labels and curation here are © What's New. Quote freely with attribution and a link back; wholesale republication of the corpus is not permitted — terms: https://whatsnew.fyi/terms. The vendors' own release notes remain their publishers'.

---

- **deprecated** — WebGL and JSEP execution providers in onnxruntime-web are deprecated in favor of the native WebGPU EP
- **added** — POSIX telemetry is now available on Linux, macOS, Android, and iOS when ONNX Runtime is built with telemetry enabled
- **removed** — Removed unused internal onnxruntime/python/tools/tensorrt dashboard tooling
- **security** — Fixed a path traversal vulnerability in TensorRT and NvTensorRTRTX engine refitting by making external-data path validation unconditional
- **security** — Validated the CPU MoE k attribute against the number of experts and fixed a CPU TensorScatter security issue
- **security** — Added missing rank, shape, and parameter validation for pooling, LSTM, DynamicQuantizeLSTM, Sampling, FeatureVectorizer, SkipLayerNorm, QLinearConv, Whisper decoding, RNN activations, GridSample, contrib Range, and CropAndResize
- **security** — Hardened CUDA indexing and buffer handling in GridSample, transpose, GatherBlockQuantized, InstanceNormalization, LayerNorm/RMSNorm, BeamSearch, DeformConv, AveragePool, and MaxPool
- **security** — Fixed packed sub-byte tensor over-copying in OrtApi::GetValue and validated DML constant tensor byte sizes
- **security** — Updated npm lockfiles, refreshed Next.js end-to-end fixture lockfile for security advisories, and upgraded adm-zip for onnxruntime-node
- **added** — Default intra-op and inter-op thread-pool sizes can now be set with ORT_INTRA_OP_NUM_THREADS and ORT_INTER_OP_NUM_THREADS environment variables
- **added** — Added weightless-model support for all initializer types and allowed zero-input EpContext nodes
- **added** — Wired maximum-shape inference into workspace estimation
- **added** — Added ONNX-domain support for rotary embedding and a fused MRotaryEmbedding contrib operator for Qwen mRoPE variants
- **added** — Added multi-shape profiling to onnxruntime_perf_test through --data_shape option
- **added** — Added verbose graph-transformer tracing and broader inference-session error-path coverage
- **added** — WebGPU now supports device-free compile-only sessions for offline graph transformation
- **changed** — Expanded CUDA plugin EP packaging and testing including Windows ARM64 package, updated package outputs, and aligned architecture selections across Python, C API, TensorRT, Node.js, and plugin packages

##### Announcements & Breaking Changes

- onnxruntime-web has announced the deprecation of WebGL and JSEP. The native WebGPU EP is the recommended path going forward. See the deprecation and migration plans for details ([#29716](https://github.com/microsoft/onnxruntime/pull/29716), [#31683](https://github.com/microsoft/onnxruntime/pull/31683)).
- POSIX telemetry is now available on Linux, macOS, Android, and iOS when ONNX Runtime is built with telemetry enabled. It does not change the public ABI, WebAssembly remains telemetry-free, and setting `ORT_DISABLE_TELEMETRY=1` before initialization disables non-Windows telemetry for the process ([#27379](https://github.com/microsoft/onnxruntime/pull/27379), [#29872](https://github.com/microsoft/onnxruntime/pull/29872)).
- The unused internal `onnxruntime/python/tools/tensorrt` dashboard tooling was removed. This does not affect the TensorRT Execution Provider APIs ([#29395](https://github.com/microsoft/onnxruntime/pull/29395)).

##### Security Fixes

###### Path, bounds, and input validation

- Fixed a path traversal vulnerability in TensorRT and NvTensorRTRTX engine refitting by making external-data path validation unconditional ([#29396](https://github.com/microsoft/onnxruntime/pull/29396)).
- Validated the CPU MoE `k` attribute against the number of experts and fixed a CPU `TensorScatter` security issue ([#29907](https://github.com/microsoft/onnxruntime/pull/29907), [#29916](https://github.com/microsoft/onnxruntime/pull/29916)).
- Added missing rank, shape, and parameter validation for pooling, LSTM and DynamicQuantizeLSTM, Sampling, FeatureVectorizer, SkipLayerNorm, QLinearConv, Whisper decoding, RNN activations, GridSample, contrib `Range`, and `CropAndResize` ([#29254](https://github.com/microsoft/onnxruntime/pull/29254), [#29255](https://github.com/microsoft/onnxruntime/pull/29255), [#29265](https://github.com/microsoft/onnxruntime/pull/29265), [#29579](https://github.com/microsoft/onnxruntime/pull/29579), [#29595](https://github.com/microsoft/onnxruntime/pull/29595), [#29605](https://github.com/microsoft/onnxruntime/pull/29605), [#29871](https://github.com/microsoft/onnxruntime/pull/29871), [#31636](https://github.com/microsoft/onnxruntime/pull/31636), [#31671](https://github.com/microsoft/onnxruntime/pull/31671), [#31675](https://github.com/microsoft/onnxruntime/pull/31675), [#31676](https://github.com/microsoft/onnxruntime/pull/31676), [#31684](https://github.com/microsoft/onnxruntime/pull/31684)).
- Hardened CUDA indexing and buffer handling in GridSample, transpose, GatherBlockQuantized, InstanceNormalization, LayerNorm/RMSNorm, BeamSearch, DeformConv, AveragePool, and MaxPool ([#29581](https://github.com/microsoft/onnxruntime/pull/29581), [#29631](https://github.com/microsoft/onnxruntime/pull/29631), [#29638](https://github.com/microsoft/onnxruntime/pull/29638), [#31640](https://github.com/microsoft/onnxruntime/pull/31640), [#31642](https://github.com/microsoft/onnxruntime/pull/31642), [#31644](https://github.com/microsoft/onnxruntime/pull/31644), [#31645](https://github.com/microsoft/onnxruntime/pull/31645), [#31647](https://github.com/microsoft/onnxruntime/pull/31647), [#31650](https://github.com/microsoft/onnxruntime/pull/31650)).
- Fixed packed sub-byte tensor over-copying in `OrtApi::GetValue` and validated DML constant tensor byte sizes ([#29157](https://github.com/microsoft/onnxruntime/pull/29157), [#31665](https://github.com/microsoft/onnxruntime/pull/31665)).

###### Supply chain and tooling

- Updated npm lockfiles, refreshed the Next.js end-to-end fixture lockfile for security advisories, and upgraded `adm-zip` for `onnxruntime-node` ([#29827](https://github.com/microsoft/onnxruntime/pull/29827), [#29926](https://github.com/microsoft/onnxruntime/pull/29926), [#31192](https://github.com/microsoft/onnxruntime/pull/31192)).

##### New Features

###### Core APIs & Runtime

- Default intra-op and inter-op thread-pool sizes can now be set with

_[Truncated at 4000 characters — full notes: https://github.com/microsoft/onnxruntime/releases/tag/v1.29.0]_