The JuliaGPU community has been a strong presence at JuliaCon for many years, and continues to be a strong foundation of Julia's overall computing ecosystem. In 2026, we propose to organize a minisymposium specifically focused on the usage and programming of GPUs (and other accelerators) in Julia. There is some overlap with an HPC minisymposium, however we intend to focus our track on very GPU-specific content or low-level details that make JuliaGPU tick. Additionally, material relating to non-GPU devices (such as TPUs, APUs, IPUs, etc.) are very welcome!
Julia has had strong GPU support for many years, starting originally with CUDAnative.jl and other CUDA-specific packages (today consolidated as CUDA.jl), and then gaining support for AMD's ROCm (AMDGPU.jl), Intel (oneAPI.jl), Metal (Metal.jl), and others for other kinds of accelerators. Additionally, we've seen the growth of GPU-specific tooling like GPUCompiler.jl and GPUArrays.jl, and the creation of unified GPU programming APIs like KernelAbstractions.jl, AcceleratedKernels.jl, and JACC.jl. Through this growth of our JuliaGPU ecosystem, the rest of Julia's ecosystems have grown in synergy, with many foundational Julia libraries now having rock-solid GPU computing support for some or all of the JuliaGPU "backends". New backends for emerging accelerators (such as TPUs, IPUs, and more) are being developed by users who wish to use them, producing a broad foundation of hardware support.
For this minisymposium, we invite talks that put a strong focus on GPU computing (whether it be about low-level GPU details, or high-level APIs for accessing GPUs), and also talks that focus on emerging or existing non-GPU accelerators, such as TPUs, IPUs, APUs, and other "XPUs". We wish to see the various backends, APIs, and integrations explored so that the community can be informed on the progress being made, and the means by which GPUs and XPUs can be utilized from Julia (or other non-Julia languages that rely on JuliaGPU).
Julian is a Research Software Engineer at MIT's JuliaLab, where he focuses on improving Julia's support for HPC and GPU computing. Julian has previously authored and maintained the AMDGPU.jl package (for programming AMD's GPUs from Julia), and now focuses his efforts on maintaining and developing the Dagger.jl package, to improve the state of productive parallel programming.