Bump Microsoft.ML.OnnxRuntime from 1.28.0 to 1.29.0

[dependabot]

Updated Microsoft.ML.OnnxRuntime from 1.28.0 to 1.29.0.

Release notes

Sourced from Microsoft.ML.OnnxRuntime's releases.

1.29.0

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, #​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, #​29872).
• The unused internal onnxruntime/python/tools/tensorrt dashboard tooling was removed. This does not affect the TensorRT Execution Provider APIs (#​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).
• Validated the CPU MoE k attribute against the number of experts and fixed a CPU TensorScatter security issue (#​29907, #​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, #​29255, #​29265, #​29579, #​29595, #​29605, #​29871, #​31636, #​31671, #​31675, #​31676, #​31684).
• Hardened CUDA indexing and buffer handling in GridSample, transpose, GatherBlockQuantized, InstanceNormalization, LayerNorm/RMSNorm, BeamSearch, DeformConv, AveragePool, and MaxPool (#​29581, #​29631, #​29638, #​31640, #​31642, #​31644, #​31645, #​31647, #​31650).
• Fixed packed sub-byte tensor over-copying in OrtApi::GetValue and validated DML constant tensor byte sizes (#​29157, #​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, #​29926, #​31192).

New Features

Core APIs & Runtime

• 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. Explicit thread settings still take precedence, and 0 preserves machine-sized defaults (#​29688).
• Added weightless-model support for all initializer types, allowed zero-input EpContext nodes, and wired maximum-shape inference into workspace estimation (#​29607, #​29799, #​31613).
• Added ONNX-domain support for rotary embedding and a fused MRotaryEmbedding contrib operator for Qwen mRoPE variants (#​29261, #​31728).
• Added multi-shape profiling to onnxruntime_perf_test through --data_shape, plus verbose graph-transformer tracing and broader inference-session error-path coverage (#​29555, #​29558, #​29569, #​29571).

Execution Provider ABI & Plugin EPs

• WebGPU now supports device-free compile-only sessions for offline graph transformation (#​29681).
• Expanded CUDA plugin EP packaging and testing, including Windows ARM64 package and size options, updated package outputs, and aligned architecture selections across Python, C API, TensorRT, Node.js, and plugin packages (#​31635, #​31722, #​31992).
• Improved plugin lifecycle handling by unloading failed EP library loads and fixing allocator-deleter lifetime (#​29634, #​29770).

Execution Provider Updates

NVIDIA CUDA EP

Attention and decoding

• Added PagedAttention with quantized KV cache, XQA decode, MLA, QK-Norm, and head-sink support (#​29912).
• Extended quantized KV-cache support with attention sinks, independent and per-channel scales, sliding-window cache support, and a fused K/V dequantization launch (#​29900, #​29904, #​31480).
• Added a cuDNN SDPA decode tier to the standard ONNX Attention CUDA kernel and enabled cuDNN SDPA for contrib Attention (#​29715, #​29717).
• Added attention_bias support to the GroupQueryAttention unfused path and state_window support to LinearAttention and CausalConvWithState for MTP (#​29525, #​31157).
• Fixed LinearAttention on GPUs with limited shared memory (#​31982).

MoE and quantized GEMM

• Added NVFP4 QMoE, including native FP4xFP4 prefill on SM120, faster decode GEMV, fused routing/finalization paths, and reduced activation and weight-dequantization overhead (#​29697, #​29824, #​29887, #​29919, #​31156, #​31159, #​31349, #​31479).
  ... (truncated)

Commits viewable in compare view.

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

---

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

• @dependabot rebase will rebase this PR
• @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
• @dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency
• @dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)

[clawsweeper]

🦞👀
ClawSweeper picked this up.

Pull request received. I will update this pull request when review starts.

[clawsweeper]

Codex review: needs maintainer review before merge. Reviewed August 17, 2026, 2:15 AM ET / 06:15 UTC.

ClawSweeper review

What this changes

The PR updates the shared .NET project's Microsoft.ML.OnnxRuntime package from 1.28.0 to 1.29.0.

Merge readiness

⚠️ Ready for maintainer review - 2 items remain

Keep open. Current main still uses 1.28.0, and this is a narrow direct production dependency update; merge after the running Windows checks and native payload compatibility evidence complete.

Priority: P2
Reviewed head: 0483434d21578aa87f726049fa38015585c8d1da

Review scores

Measure	Result	What it means
Overall readiness	🐚 platinum hermit (4/6)	The patch is tightly scoped and source-compatible on review, with native package validation remaining the material merge check.
Proof confidence	🌊 off-meta tidepool	Not applicable: This Dependabot package-only PR is exempt from contributor real-behavior proof; repository policy still calls for native payload validation before merge.
Patch quality	🐚 platinum hermit (4/6)	No actionable review findings were identified.

Verification

Check	Result	Evidence
Real behavior	Not applicable	Not applicable: This Dependabot package-only PR is exempt from contributor real-behavior proof; repository policy still calls for native payload validation before merge.
Evidence reviewed	6 items	Current main still needs the update: Current main pins Microsoft.ML.OnnxRuntime 1.28.0, while the PR head changes only that direct reference to 1.29.0. Native runtime compatibility surface: The release dependency probe loads Microsoft.ML.OnnxRuntime.dll and requires OrtEnv initialization alongside the Piper/Sherpa native stack. Established packaging protection: The repository's release test contract requires the ONNX assemblies and native DLL, and documents the supported VC++ runtime floor for onnxruntime 1.20 and later.
Findings	None	None.
Security	None	None.

How this fits together

The shared .NET project supplies managed and native audio dependencies to the Windows tray application. Its restored assets are included in released payloads, where the native Piper text-to-speech and ONNX Runtime chain is checked for successful initialization.

flowchart LR
  A[Package version] --> B[NuGet restore]
  B --> C[Shared .NET library]
  C --> D[Windows tray build]
  D --> E[Packaged Windows payload]
  E --> F[Native speech runtime check]

Loading

Before merge

• Resolve merge risk (P1) - This updates a native ONNX binary in the packaged Piper text-to-speech dependency chain; required Windows checks are still running and no PR-head packaged native-load result has been provided.
• Complete next step (P2) - Wait for required Windows CI and the native payload check; no code repair or product decision is currently needed.

Agent review details

Security

None.

Review metrics

Metric	Value	Why it matters
Dependency diff	1 file changed, 1 added, 1 removed	The PR is limited to one direct production package version change.

Merge-risk options

Maintainer options:

1. Validate the packaged native stack (recommended)
   After required checks finish, verify PR-head x64 and ARM64 payloads with the native dependency probe, including ONNX Runtime initialization where supported.

Technical review

Best possible solution:

Land the one-line update only after the PR head builds and the packaged Piper and ONNX native stack initializes for supported Windows payloads.

Do we have a high-confidence way to reproduce the issue?

Not applicable. This PR is dependency maintenance rather than a reported user failure; the native initialization path is defined in source but has not been run at the PR head.

Is this the best way to solve the issue?

Yes. Updating the existing direct reference is the narrowest maintainable path, provided packaged native-runtime compatibility is verified before merge.

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning high; reviewed against fc9add75eda7.

Labels

Label changes:

• add P2: This is routine but production-impacting native dependency maintenance for the local speech stack.
• add merge-risk: 🚨 compatibility: A new ONNX native runtime version must remain compatible with the packaged Piper/Sherpa dependency chain.
• add rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🌊 off-meta tidepool and patch quality is 🐚 platinum hermit.
• add status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Not applicable: This Dependabot package-only PR is exempt from contributor real-behavior proof; repository policy still calls for native payload validation before merge.

Label justifications:

• P2: This is routine but production-impacting native dependency maintenance for the local speech stack.
• merge-risk: 🚨 compatibility: A new ONNX native runtime version must remain compatible with the packaged Piper/Sherpa dependency chain.
• rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🌊 off-meta tidepool and patch quality is 🐚 platinum hermit.
• status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Not applicable: This Dependabot package-only PR is exempt from contributor real-behavior proof; repository policy still calls for native payload validation before merge.

Evidence

What I checked:

• Current main still needs the update: Current main pins Microsoft.ML.OnnxRuntime 1.28.0, while the PR head changes only that direct reference to 1.29.0. (src/OpenClaw.Shared/OpenClaw.Shared.csproj:24, 0483434d2157)
• Native runtime compatibility surface: The release dependency probe loads Microsoft.ML.OnnxRuntime.dll and requires OrtEnv initialization alongside the Piper/Sherpa native stack. (scripts/Test-ReleaseNativeDependencies.ps1:211, fc9add75eda7)
• Established packaging protection: The repository's release test contract requires the ONNX assemblies and native DLL, and documents the supported VC++ runtime floor for onnxruntime 1.20 and later. (tests/OpenClaw.Tray.Tests/ReleaseSigningWorkflowTests.cs:87, fc9add75eda7)
• Feature history: The current shared project and its 1.28.0 declaration were introduced in the latest release commit; the earlier local speech feature dates to the audio capability commit. (src/OpenClaw.Shared/OpenClaw.Shared.csproj:24, 4206611f2b2e)
• Release check: The current dependency declaration is included in released tag v2026.7.1; the newer PR head is not yet on main or in that release. (src/OpenClaw.Shared/OpenClaw.Shared.csproj:24, 4206611f2b2e)
• Advisory review limitation: The repository's structured autoreview helper was invoked for the PR commit but could not create its isolated temporary bundle because this review environment has no usable temporary directory; manual source, history, and diff review completed. (0483434d2157)

Likely related people:

• Régis Brid: Introduced the local Whisper and Piper speech capability that established the adjacent audio-runtime surface. (role: original audio capability contributor; confidence: high; commits: b0ba9affa25d; files: src/OpenClaw.Shared/Audio/VoiceActivityDetector.cs)
• Scott Hanselman: Introduced the current shared project file and direct ONNX Runtime package declaration on main. (role: current shared-project contributor; confidence: high; commits: 4206611f2b2e; files: src/OpenClaw.Shared/OpenClaw.Shared.csproj)
• Vincent Koc: Feature history shows repeated work on release packaging and native runtime dependencies surrounding the affected payload path. (role: native release-packaging contributor; confidence: medium; commits: 7485ce2af9da, 9e6f1caa76ad; files: scripts/Test-ReleaseNativeDependencies.ps1, src/Directory.Build.targets)

Rank-up moves

Optional improvements that raise the rating; they are not merge blockers.

• Allow required Windows CI to complete, then retain PR-head packaged ONNX initialization evidence for the supported payloads.

Rating scale

Score	Internal tier	Crab rank	Meaning
6/6	S	🦀 challenger crab	Exceptional readiness
5/6	A	🦞 diamond lobster	Very strong readiness
4/6	B	🐚 platinum hermit	Good normal PR; ordinary maintainer review
3/6	C	🦐 gold shrimp	Useful, but confidence is limited
2/6	D	🦪 silver shellfish	Proof or implementation needs work
1/6	F	🧂 unranked krab	Not merge-ready
N/A	NA	🌊 off-meta tidepool	Rating does not apply

Overall follows the weaker of proof and patch quality.
Shiny media proof means a screenshot, video, or linked artifact directly shows the changed behavior. Runtime, network, CSP, and security claims still need visible diagnostics.

Workflow

• ClawSweeper keeps one durable marker-backed review comment per issue or PR.
• Re-runs edit this comment so the latest verdict, findings, and automation markers stay together instead of adding duplicate bot comments.
• A fresh review can be triggered by eligible @clawsweeper re-review comments, exact-item GitHub events, scheduled/background review runs, or manual workflow dispatch.
• PR/issue authors and users with repository write access can comment @clawsweeper re-review or @clawsweeper re-run on an open PR or issue to request a fresh review only.
• Maintainers can also comment @clawsweeper review to request a fresh review only.
• Fresh-review commands do not start repair, autofix, rebase, CI repair, or automerge.
• Maintainer-only repair and merge flows require explicit commands such as @clawsweeper autofix, @clawsweeper automerge, @clawsweeper fix ci, or @clawsweeper address review.
• Maintainers can comment @clawsweeper explain to ask for more context, or @clawsweeper stop to stop active automation.