Issues

• [ALERT] Access to NVIDIA's Self-Hosted Runners is Expiring

  nvidia-runners
  NVIDIA/nvbench#483 · Filter by author nvidia-runner-mgmt opened on Sep 28, 2026

• Track nvbench problematic comparison cases

  NVIDIA/nvbench#481 · Filter by author fbusato opened on Sep 22, 2026

• Add a real benchmark-alteration test

  NVIDIA/nvbench#480 · Filter by author fbusato opened on Sep 22, 2026

• Add a synthetic performance alteration test

  NVIDIA/nvbench#479 · Filter by author fbusato opened on Sep 22, 2026

• Add self-comparison stability testing

  NVIDIA/nvbench#478 · Filter by author fbusato opened on Sep 22, 2026

• Create a standalone repository for nvbench evaluation tests

  NVIDIA/nvbench#477 · Filter by author fbusato opened on Sep 22, 2026

• Output of --bulk-debug-python generated by robust comparison script should include configuration values

  NVIDIA/nvbench#460 · Filter by author oleksandr-pavlyk opened on Aug 13, 2026
  

• Provide an option to reset persisting-L2 state between benchmark cells

  NVIDIA/nvbench#458 · Filter by author robobryce opened on Aug 11, 2026
  2

• Add support for int32 parameter axes

  NVIDIA/nvbench#436 · Filter by author tgujar opened on Jul 25, 2026
  3

• Add MVP instructions on addressing benchmark comparison issues

  NVIDIA/nvbench#416 · Filter by author gevtushenko opened on Jul 17, 2026
  

• NVBench instrumented benchmarks should always include bulk data in JSON output

  python
  Task

  NVIDIA/nvbench#411 · Filter by author oleksandr-pavlyk opened on Jul 16, 2026
  

• Integrate CCCL runtime APIs into nvbench

  blocked
  Task

  NVIDIA/nvbench#410 · Filter by author pciolkosz opened on Jul 13, 2026
  3

• [Skill]: Report and interpret benchmark results

  NVIDIA/nvbench#390 · Filter by author NaderAlAwar opened on Jun 9, 2026
  

• [Skill]: Choose the right NVBench timing mode

  NVIDIA/nvbench#389 · Filter by author NaderAlAwar opened on Jun 9, 2026
  

• [Skill]: Write a correct NVBench benchmark for a CUDA kernel or library call

  NVIDIA/nvbench#388 · Filter by author NaderAlAwar opened on Jun 9, 2026
  

• Add skills to enable agents to use NVBench more effectively

  NVIDIA/nvbench#387 · Filter by author NaderAlAwar opened on Jun 9, 2026
  

• Revisit handling dependent DLLs on Windows

  NVIDIA/nvbench#369 · Filter by author oleksandr-pavlyk opened on May 19, 2026
  1

• Compile the list of public API provided by NVBench library

  Task

  NVIDIA/nvbench#355 · Filter by author oleksandr-pavlyk opened on May 7, 2026

• Update examples/custom_criterion.cu

  area: examplesP2: nice to havetype: cleanup
  Task

  NVIDIA/nvbench#343 · Filter by author oleksandr-pavlyk opened on May 1, 2026
  

• Enable generation of NVBench documentation

  area: docsblocked
  Task

  NVIDIA/nvbench#324 · Filter by author oleksandr-pavlyk opened on Mar 10, 2026
  

• Extend comparison script to allow comparison of two distributions of timing samples

  Task

  NVIDIA/nvbench#313 · Filter by author jrhemstad opened on Feb 3, 2026
  1

  

• Running batch benchmarks affects cold benchmarks

  NVIDIA/nvbench#310 · Filter by author bernhardmgruber opened on Feb 2, 2026
  1

• Throughput Failed On Mutiple GPUS

  NVIDIA/nvbench#287 · Filter by author westfly opened on Nov 23, 2025
  2

• Write-only benchmark exceeds 100% bandwidth

  NVIDIA/nvbench#270 · Filter by author bernhardmgruber opened on Sep 17, 2025
  5

• Show batch time comparisons in nvbench_compare.py

  NVIDIA/nvbench#247 · Filter by author bernhardmgruber opened on Jul 17, 2025
  1