Search results

Feature Request: Expose More Core CUDA API Functions for Critical Handle Types (CUmod_st, CUfunc_st, etc.)

enhancement

help wanted

Status: Open.

#1763 In bytedeco/javacpp-presets;

· mullerhai opened on May 3, 2026

javacpp cuda nvrtcProgram + cuLaunchKernel not working correctly (CUDA 13.1 / 13.2 / Ubuntu 26.04) nvrtcProgram and launchKernel not konw the correct use style ,or maybe could not load and compile kernel function

enhancement

help wanted

Status: Open.

#1762 In bytedeco/javacpp-presets;

· mullerhai opened on May 2, 2026

Refactor ProcessGroup hierarchy for better type safety in NCCL/Gloo FSDP training

enhancement

help wanted

Status: Open.

#1761 In bytedeco/javacpp-presets;

· mullerhai opened on May 2, 2026

FSDP training causes off-heap memory continuously growing (OOM) even with PointerScope and explicit .close()

help wanted

question

Status: Open.

#1760 In bytedeco/javacpp-presets;

· mullerhai opened on May 1, 2026

should we consider adding multiple overloaded [ forward] methods to org.bytedeco.pytorch.Module？

enhancement

help wanted

Status: Open.

#1757 In bytedeco/javacpp-presets;

· mullerhai opened on Apr 26, 2026

[javacpp-pytorch] OptimizerParamGroupVector class need have push_back method to insert data

enhancement

help wanted

Status: Open.

#1756 In bytedeco/javacpp-presets;

· mullerhai opened on Apr 23, 2026

Support for Comprehensive ATen Quantized Operators (30+ Ops) for Java-based Inference

enhancement

help wanted

Status: Open.

#1745 In bytedeco/javacpp-presets;

· mullerhai opened on Mar 18, 2026

[javacpp-pytorch]do we need fix up torch.tensor(new Scalar(*) make jam crash bugs

enhancement

help wanted

Status: Open.

#1737 In bytedeco/javacpp-presets;

· mullerhai opened on Feb 18, 2026

[javacpp-PyTorch] Enhance torch::nn Options classes to support chain Fluent API (Chained Setters) similar to C++

enhancement

help wanted

Status: Open.

#1736 In bytedeco/javacpp-presets;

· mullerhai opened on Feb 12, 2026

[feature Idea for javacpp compile libtorch] Organize and arrange the code files by them Namespace

enhancement

help wanted

Status: Open.

#1735 In bytedeco/javacpp-presets;

· mullerhai opened on Feb 4, 2026

accelerating inference, let javacpp-pytorch support AOTIModelPackageLoader and AOTIModelContainerRunnerCuda and Profiler export file

enhancement

Status: Open.

#1729 In bytedeco/javacpp-presets;

· mullerhai opened on Jan 10, 2026

do we need torch.unique ,though we have torch.unique_consecutive

enhancement

help wanted

Status: Open.

#1726 In bytedeco/javacpp-presets;

· mullerhai opened on Dec 20, 2025