From Dev.to Community
Tech Articles
Curated developer articles, tutorials, and guides — auto-updated hourly
shinji shimizuMay 22, 2026 • 11 min read
HiDream Skeleton Mode: Prompt Beats OpenPose Ref — 8 Patterns Benchmarked
Benchmarking HiDream-O1-Image skeleton mode across 8 patterns reveals 3 counterintuitive findings ab...
#ai#python#machinelearning#gpu
0 0
Ingero TeamMay 26, 2026 • 4 min read
TCP Retransmits Are Not a Fabric Signal on InfiniBand
On InfiniBand the data path never touches TCP, so the retransmit proxy reads zero. The measured...
#ebpf#gpu#rdma#infiniband
0 0
Myoungho ShinMay 22, 2026 • 10 min read
Profiling a CUDA Python Program with GPUFlight
In the previous post, I used a C++ CUDA example to look at memory coalescing and how memory access.....
#performance#python#cuda#gpu
0 0
shinji shimizuMay 22, 2026 • 5 min read
Running LTX-2.3 Alongside TTS on a Single 96GB GPU with a Cold-Start Architecture
How to go from 86 GiB idle VRAM (instant OOM) to 0 GiB idle / 40 GiB peak by using a cold-start desi...
#gpu#python#machinelearning#ai
0 0
Ingero TeamMay 28, 2026 • 5 min read
Tracing torch.cuda.empty_cache() on an RTX 4090 - Where Do the 53 MB Go?
TL;DR After del tensor; torch.cuda.empty_cache(), PyTorch's caching allocator still...
#gpu#cuda#pytorch#debugging
0 0
Max VyaznikovMay 26, 2026 • 7 min read
20 Years of GPUs in Numbers: How FLOPS and TDP Grew, and Who Led the NVIDIA vs AMD Duel (+ open dataset of 13,500 GPUs)
A data-driven look at 19 years of GPUs: ~400x FP32 growth, the datacenter TDP explosion, ~100x perf/...
#gpu#machinelearning#hardware#datascience
1 0
Alan WestMay 24, 2026 • 5 min read
Why Your PyTorch Training Crawls on a Beefy GPU (And How to Fix It)
Your GPU sits at 15% utilization and bigger batches don't help? Here's how to diagnose whether you'r...
#pytorch#performance#machinelearning#gpu
0 0
shinji shimizuMay 22, 2026 • 7 min read
Cutting LTX-2 22B Peak VRAM by 40% with fp8_cast — and Why optimum-quanto Was a Trap
How fp8_cast reduced LTX-2 22B peak VRAM from 40 GiB to 24 GiB in cold-start mode, and why optimum-q...
#ai#machinelearning#gpu#python
0 0
soyMay 27, 2026 • 3 min read
CUDA 13.3 Lands, AI Writes Blackwell Kernels, & FP4 VRAM Optimization for LLMs
CUDA 13.3 Lands, AI Writes Blackwell Kernels, & FP4 VRAM Optimization for LLMs ...
#gpu#nvidia#hardware
0 0
shinji shimizuMay 22, 2026 • 5 min read
HiDream-O1-Image 3–8x Faster: Benchmarking Steps, CFG, and Resolution
Real-world timing benchmarks for HiDream-O1-Image Full — tuning steps, guidance scale, and resolutio...
#ai#machinelearning#gpu#python
0 0
soyMay 23, 2026 • 3 min read
AMD GPU/AI Launches, Legacy Driver Update & CUDA Optimization Platform
AMD GPU/AI Launches, Legacy Driver Update & CUDA Optimization Platform ...
#gpu#nvidia#hardware
0 0
Sam HosseiniMay 25, 2026 • 5 min read
How to Detect GPU Waste in a Kubernetes Cluster
GPU waste in Kubernetes does not announce itself. Your cluster shows healthy utilization. Your...
#kubernetes#gpu#mlops#devops
0 0
shinji shimizuMay 22, 2026 • 7 min read
Turning a 1-Line Idea Into a 40-Second Short with a 10-Beat Local Video Pipeline
Full pipeline: Gemma 4 31B expands a one-liner into a 10-beat script, HiDream generates images, LTX-...
#python#ai#machinelearning#gpu
0 0
shinji shimizuMay 22, 2026 • 8 min read
Five Years Later, I Finally Have 96GB VRAM — What It Actually Unlocks for Agent Loops
Not a GPU unboxing. A real look at what 96GB VRAM enables for multi-model agent pipelines — and wher...
#gpu#ai#machinelearning#python
0 0
soyMay 25, 2026 • 3 min read
PatentLLM: CUDA TileLang/Triton B200 5x Speedup, RTX 5090 Power, PTX Grammar
PatentLLM: CUDA TileLang/Triton B200 5x Speedup, RTX 5090 Power, PTX Grammar ...
#gpu#nvidia#hardware
0 0
soyMay 24, 2026 • 3 min read
RTX 5080 Undervolt Benchmarks, CGO-Free CUDA API Binding, & AMD GPU Compatibility Fix
RTX 5080 Undervolt Benchmarks, CGO-Free CUDA API Binding, & AMD GPU Compatibility...
#gpu#nvidia#hardware
0 0
soyMay 26, 2026 • 3 min read
FlashAttention CUDA Kernel, Strix Halo MOE Boost, & NVIDIA DLSS 4.5 Driver Update
FlashAttention CUDA Kernel, Strix Halo MOE Boost, & NVIDIA DLSS 4.5 Driver Update ...
#gpu#nvidia#hardware
0 0
RunC.AI OfficalMay 29, 2026 • 16 min read
5090 vs 4090 for AI Workloads: Buy, Rent, or Validate in the Cloud?
Compare 5090 vs 4090 by VRAM, bandwidth, power, and real AI workflow fit, then decide whether to buy...
#gpu#ai#cloud#hardware
0 0
cognitalkMay 28, 2026 • 1 min read
SemiAnalysis访Makora联合创始人谈自动化GPU优化与AI推理前沿
这段 GTC 研究员访谈视频由 SemiAnalysis 的 Kimbo Chen 主持，对话嘉宾是康奈尔大学助理教授、Makora（原名 Mako）的联合创始人兼首席科学官 Mohamed...
#ai#hardware#gpu#infrastructure
0 0
soyMay 22, 2026 • 4 min read
RTX 5090 Cooling, BeeLlama VRAM Opts, Resizable BAR Performance Gains
RTX 5090 Cooling, BeeLlama VRAM Opts, Resizable BAR Performance Gains Today's...
#gpu#nvidia#hardware
1 0
soyMay 28, 2026 • 3 min read
Intel Arc & Arm Mali: New GPUs, Drivers & Benchmarks for Linux
Intel Arc & Arm Mali: New GPUs, Drivers & Benchmarks for Linux Today's...
#gpu#nvidia#hardware
0 0
RunC.AI OfficalMay 29, 2026 • 10 min read
Serverless vs Dedicated VMs for GPT Endpoint Hosting: Should You Use Serverless GPU, a GPU Pod, or a VM?
Decide whether a GPT endpoint belongs on Serverless GPU, a GPU Pod, or a VM by comparing traffic sha...
#gpu#serverless#cloud#ai
0 0
RunC.AI OfficalMay 29, 2026 • 14 min read
Cost-Effective Serverless Endpoints for Docker-Based Model Inference
Build cost-effective serverless endpoints for Docker-based model inference by reducing idle GPU time...
#docker#serverless#ai#gpu
0 0
Thurmon DemichMay 29, 2026 • 7 min read
Best GPU for LTX-Video in 2026: 5 Picks (Real-Time)
RTX 4090 24GB runs LTX-Video faster than real-time. 5 GPUs ranked for image-to-video and text-to-vid...
#gpu#ltxvideo#aivideo#lightricks
0 0