Login
CUDA Ontology
(jamesakl.com) by gugagore | view | 40 comments
CUDA-l2: Surpassing cuBLAS performance for matrix multiplication through RL
(github.com) by dzign | view | 15 comments
Why CUDA translation wont unlock AMD
(eliovp.com) by JonChesterfield | view | 81 comments
Parrot – A C++ library for fused array operations using CUDA/Thrust
(nvlabs.github.io) by operator-name | view | 2 comments
Apex GPU: Run CUDA Apps on AMD GPUs Without Recompilation
(github.com) by ArchitectAI | view | 6 comments
Show HN: CUDA, Shmuda: Fold Proteins on a MacBook
(latentspacecraft.com) by geoffitect | view | 3 comments
Nvidia CUDA Tile
(developer.nvidia.com) by elkguy | view | 0 comments
Show HN: Clangd for CUDA Device Code
(docs.scale-lang.com) by JonChesterfield | view | 0 comments
Nvidia's B200: Keeping the CUDA Juggernaut Rolling Ft Verda, Formerly DataCrunch
(chipsandcheese.com) by rbanffy | view | 0 comments
Nvidia cuTile: Python DSL and a new IR for tile-based CUDA kernels
(github.com) by ashvardanian | view | 0 comments
NVIDIA CUDA Tile programming model
(developer.nvidia.com) by tanelpoder | view | 0 comments
Show HN: modal-cuda – CLI to run CUDA .cu programs on Modal GPUs
(github.com) by Sai_Praneeth | view | 0 comments
AI-Written CUDA Kernels Outperforms Nvidia's Best Matmul Library
(rohan-paul.com) by dzign | view | 0 comments
An Agent Framework with Hardware Feedback for CUDA Kernel Optimization
(arxiv.org) by PaulHoule | view | 0 comments
Understanding the CUDA Compiler and PTX with a Top-K Kernel
(blog.alpindale.net) by mfiguiere | view | 0 comments
Evaluating CUDA Tile for AI Workloads on Hopper and Blackwell GPUs
(arxiv.org) by matt_d | view | 0 comments
CUDA Released in Basic
(developer.nvidia.com) by apples2apples | view | 0 comments
Nvidia CUDA Tile
(developer.nvidia.com) by apples2apples | view | 0 comments
CUDA Tile
(techpowerup.com) by dagmx | view | 0 comments
Show HN: Free GPUs in your terminal for learning CUDA
(github.com) by RohanAdwankar | view | 0 comments
CUDA-Q Back Ends: Quantum Hardware (QPU)
(nvidia.github.io) by westurner | view | 3 comments
Nvidia CUDA Python 1.0 and CUDA 13.3 Release
(developer.nvidia.com) by ashvardanian | view | 0 comments
AutoMegaKernel: Compiling a LLM into a single CUDA kernel
(arxiv.org) by OsamaJaber | view | 0 comments
What about OpenCL and CUDA C++ alternatives?
(modular.com) by eatonphil | view | 0 comments
AutoMegaKernel: Compile an LLM into one provably-correct CUDA megakernel
(github.com) by OsamaJaber | view | 0 comments
Rl.cu: Training LLM RL with Pure CUDA
(github.com) by KJL0508 | view | 0 comments
Tiny hackable CUDA language model implementation
(github.com) by markusheimerl | view | 0 comments
Compiling Zig Kernels to CUDA PTX (Zigton Phase 1)
(portfolio.lovesahaj1225.workers.dev) by lovesahaj | view | 1 comments
How to Optimize a CUDA Matmul Kernel for cuBLAS-Like Performance: A Worklog
(siboehm.com) by Areibman | view | 0 comments
When does fragmentation occur in the CUDA caching allocator?
(docs.pytorch.org) by matt_d | view | 0 comments
Show HN: Tiny-vLLM – high performance LLM inference engine in C++ and CUDA
(github.com) by yu3zhou4 | view | 0 comments
Pushing memory bound CUDA kernels past the speed of light with data compression
(fergusfinn.com) by somnial | view | 0 comments
Nvidia CUDA 13.3 Rolls Out CUDA Python 1.0, CUDA Tile for C++
(phoronix.com) by Bender | view | 0 comments
Ternative – C++/CUDA inference engine for ternary LLMs with runtime LoRA
(github.com) by michelangeloro | view | 1 comments
Cutting inference cold starts by 40x with LP, FUSE, C/R, and CUDA-checkpoint
(modal.com) by charles_irl | view | 0 comments
CUDA Books
(github.com) by dariubs | view | 0 comments
Nvidia/Numba-CUDA-mlir: CUDA C++-style Python GPU compiler built on MLIR
(github.com) by nickwanninger | view | 0 comments
CUDA Proves Nvidia Is a Software Company
(wired.com) by bookofjoe | view | 1 comments
Atlas: An LLM inference engine written from scratch in Rust and CUDA
(atlasinference.io) by emrehan | view | 0 comments
The creators of SQL, C++, CUDA, Haskell reviewed my book on Programming History
(helloworldthebook.com) by DaleBiagio | view | 0 comments
CUDA Proves Nvidia Is a Software Company
(wired.com) by Brajeshwar | view | 0 comments
CUDA-oxide: Nvidia's official Rust to CUDA compiler
(nvlabs.github.io) by adamnemecek | view | 0 comments
Nvidia releases CUDA-Oxide 0.1 for experimental Rust-to-CUDA compiler
(phoronix.com) by birdculture | view | 0 comments
Experimental Rust-to-CUDA Compiler
(github.com) by cgravill | view | 1 comments
An experimental Rust-to-CUDA compiler from Nvidia
(nvlabs.github.io) by chenzhekl | view | 0 comments
Nvidia introduces back end for CUDA kernels in Rust
(github.com) by ketchup32613 | view | 0 comments
CUDA-oxide, a Rust-to-CUDA compiler
(nvlabs.github.io) by lacker | view | 0 comments
CUDA-oxide an experimental Rust-to-CUDA compiler
(github.com) by tzury | view | 0 comments
cuda-oxide: a custom rustc backend for compiling GPU kernels in pure Rust
(github.com) by matt_d | view | 0 comments
Cutile.jl 0.3: CUDA.jl integration, and better performance and latency
(juliagpu.org) by vchuravy | view | 0 comments