Release Notes — Version 26.10#
October 2026#
This release covers everything that landed on the 26.10 release branch since 26.09.
NuRec is now open source. The source tree and its agent skills are published at
github.com/NVIDIA/NuRec <https://github.com/NVIDIA/NuRec>_ under Apache-2.0, and the
package, module and image names change from nre to nurec as part of that move.
Most of the upgrade work below follows from the rename. See
:doc:Set Up NuRec Software <../basics/software> for how to clone the release branch and
build from source.
Upgrade From 26.09
Work through these before upgrading. Each one is detailed under Breaking Changes.
Repoint gRPC imports from
nre.grpc.protos.*tonurec.grpc.protos.*and regenerate your client stubs.Repoint container image references from
nre-gaandnre-tools-gatonurec-gaandnurec-tools-ga.Rename the
nre_toolscommand tonurec_tools, and its in-container path to/app/apps/nurec_tools.Build the gRPC client distribution from source; it is no longer published as an NGC resource.
If you build from source or extend NuRec in-process, move to PyTorch 2.11.
Relax any gRPC client that asserts an exact
grpc_api_versionpatch; it now reports0.0.5.
Breaking Changes
NuRec Is Now Open Source, and nre Is Renamed to nurec: The source tree is published under Apache-2.0 with
LICENSE.md,CONTRIBUTING.md,CODE_OF_CONDUCT.md,SECURITY.mdand GitHub issue and pull request templates, and the public tree carries abbreviated NVIDIA SPDX headers on every source file. The rename reaches every public surface:The gRPC proto packages move from
nre.grpc.protos.common,nre.grpc.protos.sensorsimandnre.grpc.protos.ipctonurec.grpc.protos.common,nurec.grpc.protos.sensorsimandnurec.grpc.protos.ipc. Generated client stubs must be regenerated against the new package names. The proto wire format is unchanged, so a client that talks to the server over the wire without importing the generated Python packages needs no change.The published container images are renamed from
nvcr.io/nvidia/nre/nre-gaandnvcr.io/nvidia/nre/nre-tools-gatonvcr.io/nvidia/nre/nurec-gaandnvcr.io/nvidia/nre/nurec-tools-ga. The registry path itself is unchanged; only the image names move.The
nre_toolscommand is renamed tonurec_tools, and its path inside the tools image moves from/app/nre/apps/nre_toolsto/app/apps/nurec_tools. The run image’s/app/runentry point is unchanged.The Python distribution is renamed from
nretonurec, and the source tree’s top-levelnre/package becomesnurec/, so in-process imports move fromnre.*tonurec.*.
The gRPC Client Distribution Is Built From Source: The gRPC client distribution was previously downloaded as an NGC resource (
nvidia/nre/nre_grpc_protos). It is now built from the published source tree with the//nurec/grpc/release:customer_distribution_tarballtarget. If you consumed the NGC resource, switch to building the tarball from a checkout of the release branch.PyTorch 2.11: The runtime moves from PyTorch 2.7 to
torch==2.11.0+cu128.nre1. This matters if you build NuRec from source, load NuRec in the same process as your own PyTorch code, or depend on the pinned Transformer Engine build, which is patched for 2.11. Reconstruction and rendering results are unchanged.NCore 19.8.0: NCore moves from 19.5.1 to 19.8.0. No serialized data format changed, so datasets and reconstructions written by earlier releases load exactly as before, but NuRec now validates sensor models more strictly — see Stricter Sensor Model Validation under Core Improvements.
gRPC API Version Is 0.0.5:
get_version()now reportsgrpc_api_version0.0.5, up from0.0.3in 26.09. Every increment so far has been additive — new discovery keys, new RPCs, and new request and response fields, all with a byte-identical proto wire format for existing clients — so a client that negotiates onmajor.minorneeds no change. A client asserting an exact patch such as(0, 0, 3)will mismatch after upgrading; relax those checks tomajor.minor.
Core Improvements
CUDA IPC Delivery for Same-Node gRPC Clients:
serve-grpccan now hand rendered frames to a client on the same node through CUDA IPC instead of copying them over the wire, behind the same gRPC endpoint. Setcuda_ipc=trueandipc_session_idon the render request to use it. The CUDA IPC path always delivers one fixed image format and supports a single session, so it is an opt-in fast path rather than a replacement for the standard delivery path. See :doc:Use the NuRec gRPC API Server <../api/grpc_api_guide>and :doc:Serve NuRec over gRPC with Multiple Processes <../api/grpc_multiprocess_serving>.Per-Request Harmonizer Control: A render request can now turn the Harmonizer on or off per request rather than only per server. Set
apply_harmonizer = trueon the request along with a validharmonizer_session_id; false or omitted leaves the request’s color unprocessed. The server manages temporal Harmonizer sessions automatically, caps concurrent sessions with--max-harmonizer-sessions(default 32), and a new history reset clears temporal state without restarting the server. The temporal Harmonizer also decodes only the current frame rather than the whole window. See :doc:Use the NuRec gRPC API Server <../api/grpc_api_guide>for the session lifecycle and its restrictions.Ego-Path Protection for Camera Ground Mesh:
export-camera-ground-meshcan now protect the drivable surface along the ego path from the mesh filters that would otherwise remove it. Pass the NCore sequence JSON with--datasetto enable it, and tune it with three new flags:--corridor-half-width-m(default 2.5) sets the half-width of the protected corridor around the ego trajectory,--corridor-z-band-m(default: off) limits protection to a height band around the trajectory and is worth setting to2.0on stacked geometry such as overpasses, and--corridor-max-edge-length-m(default 6.0) raises the long-edge limit inside the corridor.--corridor-max-edge-length-mmust exceed the mesher’s own 2.5 m edge limit, and the command rejects a smaller value rather than silently protecting nothing. Protection only retains triangles the mesher already produced; no geometry is synthesized. Exports that do not pass--datasetare unchanged.Faster Road Mesh Reconstruction: Road mesh grid reconstruction now runs on the GPU, and the grid topology defaults and the export command’s flags are documented alongside the configuration reference.
Static .ply Props as Edit Assets: Asset edits now accept a static
.plyprop as the replacement asset, alongside the asset types already supported. Rigid asset insertion is also staged atomically, so a failed insertion no longer leaves partial metadata behind.AV Benchmark Workflow Ships Its Own Configuration: The AV Benchmark workflow now carries the training preset it runs, so you can run it without an NVIDIA-internal configuration tree. The spec is
workflows/av_benchmark/specs/av_benchmark_7cam.yaml— renamed fromspecs/ci_7cam.yaml— and itsCONFIGandCONFIG_DIRvariables default to the checked-in preset atworkflows/av_benchmark/configs/. You no longer pass--set CONFIG=and--set CONFIG_DIR=to run the benchmark, and you still pass--set CONFIG=to point at a different preset. The spec’s defaults are now the benchmark’s own budgets, a 20-second clip (DURATION_SEC) and 30000 training samples (TRAIN_SAMPLES), rather than the much shorter continuous-integration budgets it previously carried, so a default run reconstructs the clip instead of smoke-testing it. The spec and its preset ship inside the released run image at those repo-relative paths, so you can run the workflow from the image without a repository checkout.A Complete AV Benchmark Chain: The workflow now runs the whole chain end to end and can optionally download its own clip, so one invocation can go from a raw PhysicalAI-AV clip to a rendered and evaluated reconstruction. It also publishes object-level metrics, the camera set and the ground mesh, reduces a clip’s metrics to a single published row, renders its publication video at 2160p, and can carry a reference USDZ’s map and ECEF pose into the exported scene package.
Per-Object PSNR: Object-level metrics now report per-object PSNR alongside the existing per-instance crops of tracked actors.
One Render-Timing Schema, Attributed to Its Scene: Render timings are unified into a single metrics file schema. The
metrics.yamlwritten by--metrics-output-dirnow files each render timing under the scene that produced it, atmetrics.per_sequence.<scene_id>.per_cameraforrender_time_ms,batch_render_time_msandbatch_render_time_per_camera_ms, and atmetrics.per_sequence.<scene_id>.per_lidarforlidar_render_time_ms. Previously all four landed together undermetrics.general, so a server holding several scenes pooled their timings into one undifferentiated set. Every timing now carries ametric_metadatablock recordingtiming_method, so you can tell wall-clock numbers from CUDA-event measurements. Metric names and units are unchanged. See :doc:Use the NuRec gRPC API Server <../api/grpc_api_guide>.Direct-Mode Render Profiling: Direct-mode rendering now reports CUDA event timings and LiDAR profiling, and backward work is labelled by the loss that owns it.
Stricter Sensor Model Validation: NCore opened its camera, LiDAR, and distortion parameter hierarchies so that you can register model classes NCore does not define. NuRec cannot render such a model, so it now reports that directly when it loads the dataset —
Unsupported NCore camera model for this repository: <type>, and the LiDAR and external-distortion equivalents — instead of failing later with an unrelated error. Datasets that use the camera and LiDAR models NCore ships are unaffected. See :doc:Prepare Data for Use with NuRec <../basics/get-data>.Atomic Artifact Publication: Reconstruction artifacts are now published atomically. Each numbered
.usdzis written to a temporary file in the artifact directory and then moved into place, andlast.usdzis hard-linked from the final numbered artifact instead of being serialized a second time. A reader can no longer observe a half-written artifact, and a failed run leaves the previously publishedlast.usdzintact. Two consequences are worth knowing if you watch or sync the artifact directory: transient files named.<filename>.<hex>.tmpappear there during a write and are removed when it finishes, andlast.usdzshares an inode with its numbered artifact rather than consuming separate disk space. Artifact contents are unchanged.Traffic Cones and Construction Barriers: Traffic cones are now initialized from LiDAR into a dedicated spherical-harmonic layer, and construction-barrier cuboids get a random initialization from upstream labels.
Rendering and Training Performance: Vehicle-light appearance lookups are cached per timestamp and their cache misses are fused and vectorized, instead of being recomputed on every layer-data call. The Slang Gaussian collector gathers each thread block’s parameters from a block-local tile rather than striding across the whole layer, improving memory locality in the forward and backward buffer operations and in the spherical-harmonic and Fourier feature copies. The collector also bounds its timestamp work and caches shutter interpolation, cuboid association is batched with bounds checks, shared rigid motion is prepared explicitly before layer collection, and several loss paths skip work they do not need — unit-weight scaling, the mean reduction for single-value losses, and the identity road-Gaussian scalar epilogue. Arithmetic, precision, tensor layouts and allocations are unchanged, so reconstruction quality is identical.
Published Agent Skills: NuRec ships agent skills in the public tree under
skills/:nurecfor reconstruction workflows,nurec-onboardfor onboarding a raw dataset end to end,physical-ai-av-benchmarkfor running one Physical AI clip through the complete AV benchmark, andphysical-ai-datasetsfor finding and downloading a Physical AI dataset. They give a coding agent the reference material and example workflows it needs to drive NuRec.physical-ai-av-benchmarkis CC-BY-4.0 AND Apache-2.0; the rest are Apache-2.0. See :doc:NuRec Agent Skills <../nurec/agent-skills>.A Dependency-Only Image for Source Builds: The published
nurec_dev_depsimage now carries only the dependency layers, on an NVIDIA Ubuntu base, rather than extending an internal development image. You build the development container yourself on top of it withdeps/docker/Dockerfile-dev.build, passing the dependency image as--build-arg BASE_IMAGE. This keeps development tooling out of the published image, which narrows both the CVE-hardening surface and the licensing surface. Building from source now requires Docker’s Buildx plugin, which Docker’s official install packages include. See :doc:Set Up NuRec Software <../basics/software>for the full command.Public Polyscope Visual Debugger: The Polyscope visual debugger is now part of the public tree.
Newer Runtime Dependencies: PyTorch moves to
2.11.0+cu128.nre1and NCore to 19.8.0, as described under Breaking Changes. The LPIPS calibration weights used by rendering metrics now come from the PyPIlpipspackage rather than a separately packaged weights archive; the weights are identical, the metric values do not change, and nothing is fetched from the network at run time. Thedecordwheel moves from0.6.0+nre3to0.6.0+nre4, which adds the LGPL notice for the ffmpeg libraries it bundles and decord’s own Apache-2.0 notice to the wheel.imageio-ffmpegis pinned at0.6.0+nre6,PyNvVideoCodecmoves to 2.2.3 for its FFmpeg 9.0.1 rebuild, the OpenCV ffmpeg build now decodes AV1,gsplatmoves past its LiDAR tiling capacity fix,ppispmoves to 1.3.0, andtransformersandurllib3move to 5.17.0 and 2.8.0. Vendored third-party code is attributed inTHIRD_PARTY_NOTICES.mdandthird_party_licenses/.
Bug Fixes
Ground Mesh Deleted Road Under the Ego Path: Fixed the camera ground mesh removing the drivable surface the ego vehicle drove over. A sparse band of points spanning the road width split the mesh into separate components, and post-processing then discarded the smaller component outright rather than filling the gap.
Spherical-Harmonic Coefficients Were Read Out of Order: Fixed
f_rest_*PLY properties being read in lexicographic rather than coefficient order, which mis-assigned higher-order spherical-harmonic coefficients when loading a PLY.LiDAR and Camera Renders Share One Execution Mode: Fixed a render failure when a LiDAR render preceded a camera render on the same scene with static-Gaussian caching enabled. The two sensor paths used different PyTorch execution modes, so the camera render then failed while updating the cache the LiDAR render had populated. Both sensors now render under the same no-grad context. Rendered values are unchanged.
Empty Tracks Crashed the Distance Mask: Fixed a CUDA failure when a track contained no instances.
Instance Segmentation Masks Were Misread: Fixed byte padding when reading instance segmentation masks.
Cameras Without Depth Broke Point Cloud Initialization: Camera-based point cloud initialization now skips cameras that have no depth rather than failing.
Cuboid Labels Were Matched to the Wrong Frames: Cuboid labels are now matched to camera frames per track.
Ground Mesh Initialization Was Not Repeatable: Fixed ground mesh initialization producing different results across runs with the same inputs.
Deform Tables Did Not Grow With Inserted Tracks: Fixed the deformation tables not growing when a track is inserted, and added a topology preflight that rejects an unsafe table deformation rather than corrupting it.
Failed Asset Insertions Left Metadata Behind: A failed asset insertion now rolls back its metadata.
Checkpoint RNG Restore: Fixed the restored CPU random number generator being shifted by loader creation, and restored the default random streams on checkpoint load. The RNG invariants are now documented.
CUDA Compatibility Libraries Lost Their SONAME Symlinks: Fixed the packaged CUDA compatibility libraries dropping their SONAME symlinks.
nurec_tools Masked Import Failures:
nurec_toolsnow fails when an enabled tool cannot be imported, rather than continuing without it.
Known Issues
dynamic_objectsreplaces the actor set for a request rather than overriding the actors it lists: A non-emptydynamic_objectslist is the complete actor set for that render. The server draws the tracks you list and no others, so every recorded actor you leave out is absent from the frame, and the response still carries transport statusOK. A client that sends only the actor it moved loses every other actor with no error. Send every actor whose trajectory overlaps the frame window on every request, and substitute your ownpose_paironly for the actors you drive. An empty list is the separate instruction to draw all recorded actors at their captured poses. This release has no override-only mode. See :doc:gRPC API Guide <../api/grpc_api_guide>.get_dynamic_objectsreturns an out-of-order first timestamp for actors that appear before the scene starts: When an actor’s first recorded pose predates the scene start, its negative microsecond offset wraps in the unsignedtimestamp_usfield and is returned as a value near2**64, leaving the trajectory out of time order. One Physical AI scene showed this for 6 of its 30 actors. Detect it by testingtimestamp_us > 2**63or by checking that timestamps strictly increase;timestamp_us - 2**64recovers the original offset, so the affected entry can be converted and re-sorted rather than discarded. Rendering is unaffected, becauserender_cameraandrender_lidarplace actors from theDynamicObject.pose_pairsupplied for every actor the request lists. An actor the request omits is not drawn at all. A fix is targeted for 26.11. See :doc:gRPC API Guide <../api/grpc_api_guide>.render_cameraaccepts a nonfinite sensor pose and returns a distorted image with status OK: Setting anysensor_posecomponent toNaNor an infinity produces a badly smeared render reported at transport statusOK, so the status code alone does not indicate a usable image. Validate that every pose component is finite before sending the request. Server state is unaffected: a subsequent valid request returns the expected image with no restart. A fix that rejects nonfinite pose values withINVALID_ARGUMENTis targeted for 26.11. See :doc:gRPC API Guide <../api/grpc_api_guide>.General Linux aarch64 is not a supported platform: The supported user platform for NuRec remains Linux x86_64. The published source release pins x86_64 dependencies only, and Grace-Blackwell enablement reaches you only through the
-arm64or-multiarchcontainer tags. Those image builds do not run on release-branch pipelines by default, so an-arm64or-multiarchtag for a given 26.10 build may not exist unless it was triggered manually. See :doc:Hardware Setup and Requirements <../basics/hardware>.