Hackernews posts about Wgpu
- Rust running on every GPU (rust-gpu.github.io)
- Writing a Rust GPU kernel driver: a brief introduction on how GPU drivers work (www.collabora.com)
- Running GPT-OSS-120B at 500 tokens per second on Nvidia GPUs (www.baseten.co)
- ARM adds neural accelerators to GPUs (newsroom.arm.com)
- A GPU Calculator That Helps Calculate What GPU to Use (calculator.inference.ai)
- Investigating the Nvidia AI GPU Black Market [video] (www.youtube.com)
- Rust Running on Every GPU (rust-gpu.github.io)
- One GPU translates into three to five of the fastest Ethernet switch ports (www.theregister.com)
- Scale: Natively compile CUDA applications for AMD GPUs (docs.scale-lang.com)
- An Introduction to GPU Profiling and Optimization (www.bitsand.cloud)
- GPU Memory Snapshots: fast container cold boots (modal.com)
- Show HN: Selfhostllm.org – Plan GPU capacity for self-hosting LLMs (selfhostllm.org)
- Arm juices mobile GPUs with neural tech for better graphics (www.theregister.com)
- Legendary GPU architect Raja Koduri's startup leverages RISC-V and targets CUDA (www.tomshardware.com)
- GPUs go brrr with Mojo: Algorithms (shubhamg.in)
- AMD's AI datacenter revenue dived due to US ban on China GPU exports (www.theregister.com)
- Descend: A Safe GPU Systems ProgrammingLanguage [pdf] (steuwer.info)
- A parallel path for GPU restore in CRIU (lwn.net)
- PyTorch in One Hour: From Tensors to Training Neural Networks on Multiple GPUs (sebastianraschka.com)
- The Nvidia AI GPU black market: smuggling, arrests, & high-tech crime (store.gamersnexus.net)
- ARMing GPUs: On the Memory Subsystem of Grace Hopper GH200 [video] (www.youtube.com)
- How to Think About GPUs (jax-ml.github.io)