My First Multi-GPU Kernel: Writing All-to-All for AMD MI300X
gau-nernst.github.io·2d·
Discuss: Hacker News
🧮Compute shaders
Flag this post
On the Structure of Floating-Point Noise in Batch-Invariant GPU Matrix Multiplication
arxiv.org·1d
🧩Procedural Generation
Flag this post
Cutting LLM Batch Inference Time in Half: Dynamic Prefix Bucketing at Scale
daft.ai·13h·
Discuss: Hacker News
🔍Algorithms
Flag this post
Inline vs. Pipeline Ray Tracing
evolvebenchmark.com·16h·
Discuss: Hacker News
🖼️Rendering
Flag this post
Ranking LLMs based on 180k French votes (French government's AI arena)
comparia.beta.gouv.fr·17h·
Discuss: Hacker News
📐Computational geometry
Flag this post
A hitchhiker's guide to CUDA programming
seanzhang.me·5d·
Discuss: Hacker News
🧮Compute shaders
Flag this post
A Soft‑Fork Proposal for Blockchain‑Based Distributed AI Computation
hackernoon.com·1d
🔍Algorithms
Flag this post
Real-time stock volatility prediction with deep learning on a time-series DB
medium.com·22h·
Discuss: Hacker News
🧱Data structures
Flag this post
Enabling Trillion-Parameter Models on AWS EFA
research.perplexity.ai·6h·
Discuss: Hacker News
🧩Procedural Generation
Flag this post
[Talk] Improving the Incremental System in the Rust Compiler
blog.goose.love·10h
🦀Rust
Flag this post
Geonum – geometric number library for unlimited dimensions with O(1) complexity
github.com·1d·
Discuss: Hacker News
📐Computational geometry
Flag this post
Why is AI Generated Rust slow when compared with Go/C#/Node/JavaScript
srid68.github.io·15h·
Discuss: Hacker News
🦀Rust
Flag this post
Video Invisible Watermarking at Scale
engineering.fb.com·12h·
Discuss: Hacker News
🧩Procedural Generation
Flag this post
Pixel Art Filtering
jorenjoestar.github.io·14h·
Discuss: Hacker News
🧮Compute shaders
Flag this post
The Evolution of GPUs: How Floating-Point Changed Computing
dell.com·2d·
Discuss: Hacker News
🏗️Computer-aided design
Flag this post
Can-t stop till you get enough
cant.bearblog.dev·2d·
Discuss: Hacker News
💻Programming
Flag this post
Nvidia, Deutsche Telekom strike €1B partnership for a data center in Munich
techcrunch.com·16h·
Discuss: Hacker News
🧩Procedural Generation
Flag this post
Topographical sparse mapping: A training framework for deep learning models
sciencedirect.com·9h·
Discuss: Hacker News
🧩Procedural Generation
Flag this post
Design of quasi phase matching crystal based on differential gray wolf algorithm
arxiv.org·1d
📐Computational geometry
Flag this post
User Experience Is Computation
madebynathan.com·3h·
Discuss: Hacker News
🧩Procedural Generation
Flag this post