From Dev.to Community
Tech Articles
Curated developer articles, tutorials, and guides � auto-updated hourly
Ingero Team3d ago • 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
Ingero Team7h ago • 6 min read
From Kernel Scheduler to Python Source Line: Tracing a GPU Stall End to End
TL;DR A GPU that reports 97% utilization can still be the slowest part of a training step,...
#ebpf#gpu#python#observability
0 0
Max Vyaznikov3d ago • 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
Ingero Team1d ago • 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
Alan West4d ago • 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
soy1d ago • 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
Sam Hosseini4d ago • 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
soy2d ago • 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
soy3d ago • 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
soy4d ago • 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
soy5d ago • 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
RunC.AI Offical16h ago • 15 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
cognitalk1d ago • 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
soy6d ago • 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
soy22h ago • 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 Offical16h ago • 9 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 Offical16h ago • 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 Demich3d ago • 7 min read
Best GPU for Mistral Models in 2026 (5 Picks Ranked)
Best GPU picks for running Mistral 7B, Mixtral 8x7B, and Mistral Large locally. VRAM needs, speed be...
#gpu#mistral#mixtral#llm
0 0
RunC.AI Offical16h ago • 18 min read
Best 7 Cloud GPU Platforms for TensorFlow Training
Compare the best cloud GPU platforms for TensorFlow training by cost, GPU tiers, storage fit, and wh...
#gpu#tensorflow#cloud#ai
0 0
Thurmon Demich5d ago • 5 min read
Best GPU for Llama 4 Scout (109B MoE) in 2026 Ranked
Best GPUs for running Llama 4 Scout (109B MoE, 17B active) locally in 2026 — VRAM needs, quantizatio...
#gpu#llama4#scout#llm
0 0
Thurmon Demich6d ago • 4 min read
How Much VRAM Do You Need for Flux? (2026 Guide)
Exact VRAM requirements for Flux image generation. Schnell, dev, ControlNet, fine-tuning -- every wo...
#vram#flux#imagegeneration#gpu
0 0
Thurmon Demich1d ago • 3 min read
Best GPU for LLM Fine-Tuning in 2026 (Ranked Picks)
Best GPUs for LoRA, QLoRA, and full fine-tuning of LLMs. VRAM requirements, speed benchmarks, and pr...
#gpu#finetuning#lora#qlora
0 0
Cyfuture AI1d ago • 5 min read
Liquid Cooled AI Data Centers: The Best Solution for GPU-Intensive AI Applications
Artificial Intelligence (AI) is transforming industries worldwide, from healthcare and finance to...
#ai#datacenter#gpu#webdev
0 0
Thurmon Demich4d ago • 6 min read
Best GPU for Flux.2 in 2026: 5 Cards Ranked (FP8 Ready)
RTX 5080 16GB now runs Flux.2 32B thanks to NVIDIA's FP8 path. 5 GPUs ranked for Flux.2 in 2026 — VR...
#gpu#flux2#flux#imagegeneration
0 0