DEV Community

# gpu

Posts

👋 Sign in for the ability to sort posts by relevant, latest, or top.
Intel Xe3P Leaks 160GB LPDDR5X; FlashAttention-2 in CuTe & Custom CUDA GPT-2 Engine
soytuber profile
soy

Intel Xe3P Leaks 160GB LPDDR5X; FlashAttention-2 in CuTe & Custom CUDA GPT-2 Engine

#gpu #nvidia #hardware
Comments
3 min read
Building llama.cpp from source on a Dell Precision T5820 with an RTX 3090 Ti (after seven power cycles)
ianlpaterson profile

Building llama.cpp from source on a Dell Precision T5820 with an RTX 3090 Ti (after seven power cycles)

#ai #llm #gpu #linux
Comments
16 min read
Three Months of Speed-Up Experiments on a 3090 Ti: Autoregressive DFlash MTP for Qwen3.6-27B
ianlpaterson profile

Three Months of Speed-Up Experiments on a 3090 Ti: Autoregressive DFlash MTP for Qwen3.6-27B

#ai #llm #gpu #performance
Comments
18 min read
Why MTP doesn't speed up your llama.cpp inference (and how to actually fix it)
alanwest profile

Why MTP doesn't speed up your llama.cpp inference (and how to actually fix it)

#llm #performance #machinelearning #gpu
Comments
5 min read
267 tok/s local inference on RTX 5090 – llama.cpp MTP + Qwen3-35B-A3B MoE
prospectorlabs profile
gen

267 tok/s local inference on RTX 5090 – llama.cpp MTP + Qwen3-35B-A3B MoE

#llm #machinelearning #llama #gpu
Comments
1 min read
GPU Bottleneck Analyzer, NVIDIA Rubin VRAM Demands, and Qwen VRAM Optimization
soytuber profile
soy

GPU Bottleneck Analyzer, NVIDIA Rubin VRAM Demands, and Qwen VRAM Optimization

#gpu #nvidia #hardware
1
Comments
4 min read
Production-Ready GPU Inference Autoscaling on EKS with Karpenter, KEDA, and Dragonfly
tazmainiandevil profile

Production-Ready GPU Inference Autoscaling on EKS with Karpenter, KEDA, and Dragonfly

#kubernetes #aws #devops #gpu
Comments
26 min read
GPU Hardware & Driver Update: RTX 5090 Benchmarks, llama.cpp MTP, Windows 11 Fix
soytuber profile
soy

GPU Hardware & Driver Update: RTX 5090 Benchmarks, llama.cpp MTP, Windows 11 Fix

#gpu #nvidia #hardware
Comments
3 min read
CUDA Cutile-rs Beta, AMD FSR 4.1 Release, & Forza Horizon 6 GPU Benchmarks
soytuber profile
soy

CUDA Cutile-rs Beta, AMD FSR 4.1 Release, & Forza Horizon 6 GPU Benchmarks

#gpu #nvidia #hardware
Comments
3 min read
Same eBPF, Different Vendor: Tracing libhip Calls on AMD ROCm
ingero profile

Same eBPF, Different Vendor: Tracing libhip Calls on AMD ROCm

#linux #programming #gpu #performance
Comments
3 min read
Best GPU for Llama 70B in 2026 (48GB+ VRAM Required)
thurmon_demich profile

Best GPU for Llama 70B in 2026 (48GB+ VRAM Required)

#gpu #llama #70b #vram
Comments
6 min read
From TCP Retransmits to MCP-Driven Cluster Investigations: An eBPF GPU Agent Retrospective
ingero profile

From TCP Retransmits to MCP-Driven Cluster Investigations: An eBPF GPU Agent Retrospective

#ebpf #gpu #observability #mcp
1
Comments
8 min read
From Zero to Supercomputing: A Beginner-Friendly Guide to Using HPC Clusters Like CINECA
kenryikegbo profile

From Zero to Supercomputing: A Beginner-Friendly Guide to Using HPC Clusters Like CINECA

#gpu #cluster #supercomputers #ai
Comments
5 min read
What Inference-Platform Benchmark Posts Leave Out
ingero profile

What Inference-Platform Benchmark Posts Leave Out

#machinelearning #ai #gpu #performance
Comments
8 min read
Why CUDA kernels silently corrupt memory and how to catch the bug
alanwest profile

Why CUDA kernels silently corrupt memory and how to catch the bug

#cuda #rust #debugging #gpu
Comments
5 min read
👋 Sign in for the ability to sort posts by relevant, latest, or top.
HTTPS ¡ dev.to
← Home