Hacker News with Generative AI: GPU

EGPU: Extending eBPF Programmability and Observability to GPUs (aptaracorp.com)
Precise GPU observability and programmability are essential for optimizing performance in AI workloads and other computationally intensive high-performance computing (HPC) applications.
GPU, Programming, Observability, AI
15 points by tanelpoder 16 days ago | 4 comments
GPU Server with 8 RTX 4090 (a16z.com)
In today’s AI-driven world, the ability to train AI models locally and perform fast inference on GPUs at an optimal cost is more important than ever.
GPU, AI, Hardware, Training, Inference
5 points by m3at 21 days ago | 2 comments
The Asus Ascent GX10 a Nvidia GB10 Mini PC with 128GB of Memory and 200GbE (servethehome.com)
NVIDIA’s platform, previously codenamed Project DIGITS, is a hit at GTC 2025. Apparently, big customers are asking if they can get a DGX Spark thrown in with large GPU purchases. The reason is simple, this is a mini PC form factor that packs an Arm CPU and a Blackwell GPU that are co-packaged, a 128GB LPDDR5x shared memory, multiple ports of USB4, and even a ConnectX-7 NIC for 200GbE clustering.
Mini PCs, NVIDIA, Arm, GPU, Networking
46 points by rbanffy 36 days ago | 50 comments
AMD Radeon RX 9070 Series Linux GPU Compute Performance (phoronix.com)
In addition to the Radeon RX 9070 series Linux gaming/graphics benchmarks with today's embargo lift, I've also spent some time working on some GPU compute benchmarks for these first RDNA4 graphics cards.
AMD, Linux, GPU, Graphics Cards, Performance
13 points by mfiguiere 51 days ago | 1 comments
DeepSeek open source DeepEP – library for MoE training and Inference (github.com/deepseek-ai)
DeepEP is a communication library tailored for Mixture-of-Experts (MoE) and expert parallelism (EP). It provides high-throughput and low-latency all-to-all GPU kernels, which are also as known as MoE dispatch and combine. The library also supports low-precision operations, including FP8.
Machine Learning, Open Source, Software, Deep Learning, GPU
536 points by helloericsf 60 days ago | 71 comments
DeepSeek Open Source FlashMLA – MLA Decoding Kernel for Hopper GPUs (github.com/deepseek-ai)
FlashMLA is an efficient MLA decoding kernel for Hopper GPUs, optimized for variable-length sequences serving.
GPU, Open Source, Machine Learning, Decoding, Computer Vision
441 points by helloericsf 61 days ago | 108 comments
The Ultra-Scale Playbook: Training LLMs on GPU Clusters (huggingface.co)
Refreshing
Machine Learning, GPU, Training, AI
33 points by jxmorris12 65 days ago | 3 comments
Ask HN: Confused about how DeepSeek hurts Nvidia (ycombinator.com)
I’m genuinely confused about why people think Deepseeks results will mean fewer GPUs being needed in the future.
Artificial Intelligence, GPU, Hardware
36 points by prng2021 88 days ago | 43 comments
Bilinear down/upsampling, aligning pixel grids, and that infamous GPU half pixel (2021) (bartwronski.com)
See this ugly pixel shift when upsampling a downsampled image? My post describes where it can come from and how to avoid those!
Image Processing, Computer Graphics, GPU, Pixel Manipulation, Downsampling/Upsampling
136 points by fanf2 89 days ago | 23 comments
Show HN: A GPU-accelerated MD5 Hash Cracker, Written Using Rust and CUDA (github.com/vaktibabat)
MD5 hash cracking with CUDA and Rust, implemented from scratch
Security, Programming, Rust, GPU, CUDA
4 points by vaktibabat 116 days ago | 0 comments
Chinese GPU designers received key technologies from British company Imagination (tomshardware.com)
Chinese Companies, GPU, Technology, Semiconductors
15 points by ksec 127 days ago | 2 comments
GPU Glossary (modal.com)
GPU, Glossary, Graphics, Hardware, Technology
5 points by georgehill 128 days ago | 0 comments
Show HN: Svader – Create GPU-rendered Svelte components (github.com/sockmaster27)
Create GPU-rendered Svelte components with WebGL and WebGPU fragment shaders.
Web Development, Svelte, GPU, Graphics, WebGPU
189 points by sokmastr 133 days ago | 41 comments
GPU Glossary (modal.com)
We wrote this glossary to solve a problem we ran into working with GPUs here at Modal : the documentation is fragmented, making it difficult to connect concepts at different levels of the stack, like Streaming Multiprocessor Architecture , Compute Capability , and nvcc compiler flags .
GPU, Glossary, Hardware, Computer Science, Documentation
15 points by iamwil 134 days ago | 1 comments
Exploring inference memory saturation effect: H100 vs. MI300x (dstack.ai)
GPU memory plays a critical role in LLM inference, affecting both performance and cost. This benchmark evaluates memory saturation’s impact on inference using NVIDIA's H100 and AMD's MI300x with Llama 3.1 405B FP8.
GPU, Benchmarking, Inference, Hardware
57 points by latchkey 141 days ago | 12 comments
Compilation on the GPU? A Feasibility Study (dl.acm.org)
GPU, Compilation, Feasibility Study, Performance
27 points by luu 143 days ago | 13 comments
Compilation on the GPU? A Feasibility Study (2022) (dl.acm.org)
The emergence of highly parallel architectures has led to a renewed interest in parallel compilation.
GPU, Compilation, Feasibility Study, Computer Science
5 points by fanf2 144 days ago | 0 comments
Scale (run CUDA on AMD GPUs without mods) supports gfx900 and gfx1102 (scale-lang.com)
GPU, Open Source, Programming, AMD
28 points by mgl 145 days ago | 2 comments
Optimizing a Rust GPU matmul kernel (rust-gpu.github.io)
I read the excellent post Optimizing a WebGPU Matmul Kernel for 1TFLOP+ Performance by Zach Nussbaum and thought it might be fun to reimplement it with Rust GPU.
Rust, GPU, Performance Optimization, WebGPU
97 points by lukastyrychtr 147 days ago | 12 comments
Optimizing a Rust GPU matmul kernel (rust-gpu.github.io)
I read the excellent post Optimizing a WebGPU Matmul Kernel for 1TFLOP+ Performance by Zach Nussbaum and thought it might be fun to reimplement it with Rust GPU.
Rust, GPU, Performance Optimization
5 points by LegNeato 152 days ago | 0 comments
$2 H100s: How the GPU Rental Bubble Burst (latent.space)
H100s used to be $8/hr if you could get them. Now there's 7 different places sometimes selling them under $2. What happened?
GPU, Cloud Computing, Market Trends, Hardware
403 points by swyx 197 days ago | 279 comments
Scuda – Virtual GPU over IP (github.com/kevmo314)
SCUDA is a GPU over IP bridge allowing GPUs on remote machines to be attached to CPU-only machines.
Remote Computing, Virtualization, GPU, Software, Networking
207 points by kevmo314 199 days ago | 40 comments
Show HN: Squey, an open-source GPU-accelerated data visualization software (squey.org)
Squey 5.0 is out! Check out the new Parquet plugin and the revamped UISquey
Open Source, Data Visualization, Software, GPU
66 points by jbleonesio 200 days ago | 13 comments
炊紙(kashikishi) is a text editor that utilizes GPU to edit text in a 3D space (github.com/mitoma)
炊紙は三次元空間上でテキストを編集できるテキストエディタです。「かしきし」と発音します。
Text Editors, GPU, 3D, Software, Programming
240 points by hiroshi3110 204 days ago | 101 comments
Hetzner introduces GPU server for AI training (hetzner.com)
Discover the next level of performance with the new GEX130 dedicated GPU server. Equipped with the NVIDIA RTX™ 6000 Ada generation graphics card, it can put ideas into practice even faster and highly complex tasks can be completed efficiently.
AI, Cloud Computing, Hardware, Server, GPU
75 points by rene_d 213 days ago | 52 comments
Writing Portable Rendering Code with Nvrhi (nvidia.com)
Modern graphics APIs, such as Direct3D 12 and Vulkan, are designed to provide relatively low-level access to the GPU and eliminate the GPU driver overhead associated with API translation.
Graphics Programming, Game Development, GPU, Rendering
30 points by Flex247A 216 days ago | 7 comments
Show HN: Oblivus GPU Cloud – On-Demand H100s from $1.98/hr – $25 Free Credit (oblivus.com)
Democratized GPU Cloud starting at only $0.12/hr! No quotas, no restrictions.
Cloud Computing, GPU, Pricing, Machine Learning, Free Trials
23 points by oblivuslimited 220 days ago | 25 comments
GPU Debug Scopes (wunkolo.github.io)
Rendering APIs these days tend to capture their gpu workloads into a serialized form such as a command-buffer or command-list to be dispatched at a later time into a work-queue.
Graphics Programming, GPU, Debugging, Rendering APIs
102 points by qsantos 221 days ago | 7 comments
Show HN: Attaching to a virtual GPU over TCP (thundercompute.com)
Virtualization, Networking, GPU, Cloud Computing
344 points by bmodel 259 days ago | 108 comments
Open-Source AMD GPU Implementation of CUDA "Zluda" Has Been Taken Down (phoronix.com)
Open Source, AMD, GPU, CUDA
39 points by TiredOfLife 263 days ago | 16 comments