DEV Community

#gpu

Posts

đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.
Inside SFPU Overflow Bugs: How a 40-Year-Old Rounding Trick Breaks on Modern AI Accelerators

Inside SFPU Overflow Bugs: How a 40-Year-Old Rounding Trick Breaks on Modern AI Accelerators

Comments
3 min read
I Put a Paid AI Video Generator on My Own Gaming GPU — No Cloud Bill

I Put a Paid AI Video Generator on My Own Gaming GPU — No Cloud Bill

Comments 1
3 min read
Eggs, Cholesterol, and GPU Flags

Eggs, Cholesterol, and GPU Flags

Comments
3 min read
Reverse-Engineering NVIDIA: Modifying a CUDA binary

Reverse-Engineering NVIDIA: Modifying a CUDA binary

Comments
4 min read
From API to GPU, Week 6 (Part 2): Watching a Neural Network Learn

From API to GPU, Week 6 (Part 2): Watching a Neural Network Learn

Comments
17 min read
From API to GPU, Week 6 (Part 1): A Model That Predicts, and How Wrong It Is

From API to GPU, Week 6 (Part 1): A Model That Predicts, and How Wrong It Is

Comments
13 min read
Scale Before the Spike: Predictive Autoscaling for GPU Workloads on Kubernetes

Scale Before the Spike: Predictive Autoscaling for GPU Workloads on Kubernetes

Comments
6 min read
DGX Spark (GB10) memory sizing for LLM serving: the numbers

DGX Spark (GB10) memory sizing for LLM serving: the numbers

Comments
7 min read
How Many AI Avatars Can One GPU Handle? Real-World Test Reveals 4 Avatars at ÂĄ7,600 Each per Month

How Many AI Avatars Can One GPU Handle? Real-World Test Reveals 4 Avatars at ÂĄ7,600 Each per Month

Comments
9 min read
OpenSearch Service GPU Acceleration and Auto-Optimization: vector search with ease

OpenSearch Service GPU Acceleration and Auto-Optimization: vector search with ease

1
Comments
4 min read
I Tried Getting Closer to the GPU With Triton

I Tried Getting Closer to the GPU With Triton

1
Comments
6 min read
Why Chromium Was Ignoring My GPU — And How I Boosted Performance from 4fps to 58fps

Why Chromium Was Ignoring My GPU — And How I Boosted Performance from 4fps to 58fps

Comments
10 min read
Borrowed an H100 but couldn't draw a single frame — Why compute GPUs and rendering GPUs are different beasts

Borrowed an H100 but couldn't draw a single frame — Why compute GPUs and rendering GPUs are different beasts

Comments
6 min read
Pure JAX on G5g: Serving Gemma 4 on Graviton and a T4G

Pure JAX on G5g: Serving Gemma 4 on Graviton and a T4G

Comments
8 min read
I Created a 24/7 AI Avatar That Streams Without Human Intervention — Only 'Verification,' 'Eyes,' and 'Ears' Remain for Humans

I Created a 24/7 AI Avatar That Streams Without Human Intervention — Only 'Verification,' 'Eyes,' and 'Ears' Remain for Humans

1
Comments
9 min read
đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.