Issues

• Occupancy 有支持吗？

  NVIDIA-AI-IOT/Lidar_AI_Solution#373 · Filter by author tianle-BigRice opened on Jul 31, 2026

• Need Support for libspconv.so without gridsize limitation

  NVIDIA-AI-IOT/Lidar_AI_Solution#372 · Filter by author sddft opened on Jul 9, 2026

• Request libspconv.so for Jetson Orin JetPack 6 CUDA 12.x with sm_87

  NVIDIA-AI-IOT/Lidar_AI_Solution#370 · Filter by author Xiaosong020 opened on Jun 9, 2026
  1

• Assert failed 💀. false in file src/spconv/implicit-gemm.cu:383, message: Can not find any matched kernel 128 x 64

  NVIDIA-AI-IOT/Lidar_AI_Solution#369 · Filter by author qinjunji opened on May 19, 2026
  1

• Add operation to engine.hpp

  NVIDIA-AI-IOT/Lidar_AI_Solution#368 · Filter by author Mysterium-sch opened on Apr 30, 2026

• ValueError：Negative values in amax

  NVIDIA-AI-IOT/Lidar_AI_Solution#367 · Filter by author songyonger opened on Mar 19, 2026

• camera geometry in bevfusion has problem

  NVIDIA-AI-IOT/Lidar_AI_Solution#366 · Filter by author LIUHAO121 opened on Mar 13, 2026

• [Bug] Engine build segfaults when SparseConvolution layer exceeds 128 channels (v1.3.2)

  NVIDIA-AI-IOT/Lidar_AI_Solution#363 · Filter by author kkurang opened on Mar 10, 2026
  2

• [Request] Support voxelnext to onnx conversion

  NVIDIA-AI-IOT/Lidar_AI_Solution#362 · Filter by author Mysterium-sch opened on Mar 8, 2026
  1

• libspconv lost support of cuda sm87

  NVIDIA-AI-IOT/Lidar_AI_Solution#360 · Filter by author NoahCSHN opened on Mar 2, 2026
  1

• Is the new version of the sparse convolution library for x86, updated in December 2025, damaged?

  NVIDIA-AI-IOT/Lidar_AI_Solution#359 · Filter by author qichen2006 opened on Feb 13, 2026
  1

• Can you provide all the input dimension types supported by spconv?

  NVIDIA-AI-IOT/Lidar_AI_Solution#358 · Filter by author chjXu opened on Feb 3, 2026

• libspconv.so.1.2.1 supports CUDA Graph question

  NVIDIA-AI-IOT/Lidar_AI_Solution#356 · Filter by author Muxindawang opened on Jan 13, 2026
  1

• Request for arch64_cuda13.0 sm110 Support

  NVIDIA-AI-IOT/Lidar_AI_Solution#352 · Filter by author lubangnshi opened on Dec 8, 2025
  2

• Camera placement and Choice

  NVIDIA-AI-IOT/Lidar_AI_Solution#351 · Filter by author rvxfahim opened on Nov 23, 2025

• I ran into a problem when deploying this project on Orin.

  NVIDIA-AI-IOT/Lidar_AI_Solution#350 · Filter by author zjh1378805302 opened on Nov 20, 2025
  5

• evaluation metrics was very bad

  NVIDIA-AI-IOT/Lidar_AI_Solution#348 · Filter by author dong-hub-png opened on Nov 12, 2025

• How to Perform CUDA Quantization for a Vision-Only BEVFusion Model?

  NVIDIA-AI-IOT/Lidar_AI_Solution#346 · Filter by author bbshdh56 opened on Oct 28, 2025

• Assert failed 💀. false in file src/spconv/implicit-gemm.cu:383, message: Can not find any matched kernel 128 x 256 Aborted (core dumped)

  NVIDIA-AI-IOT/Lidar_AI_Solution#345 · Filter by author Xiao-Hu-Z opened on Oct 21, 2025
  1

• SparseCon2d 部署是否支持

  NVIDIA-AI-IOT/Lidar_AI_Solution#344 · Filter by author Xiao-Hu-Z opened on Oct 13, 2025
  1

• load nuscenes mini dataset ，the result seems incorrect，but when i load the sample ,it is correct ,so strange

  NVIDIA-AI-IOT/Lidar_AI_Solution#343 · Filter by author maokang94 opened on Oct 6, 2025
  1

• What Does the 1×5 Input Dimension Represent in a LiDAR Backbone Model? BEVFusion

  NVIDIA-AI-IOT/Lidar_AI_Solution#342 · Filter by author dheepak-gan opened on Sep 25, 2025
  2

• spconv-v1.1.10 for aarch64 with CUDA-12.x

  NVIDIA-AI-IOT/Lidar_AI_Solution#340 · Filter by author shubhendu-ranadive opened on Sep 3, 2025
  1

• when eval the model result，the bbox AP is all zero,how to fix it

  NVIDIA-AI-IOT/Lidar_AI_Solution#339 · Filter by author knavezl opened on Aug 26, 2025

• Voxel count limit warning in spconv 1.2.1 - need configurable maximum voxel threshold

  NVIDIA-AI-IOT/Lidar_AI_Solution#338 · Filter by author LexcaliburR opened on Aug 22, 2025