---
title: "#amd articles | Hashnode"
description: "Explore 51 articles tagged #amd on Hashnode. Tutorials, guides, and how-tos from the community."
url: "https://hashnode.com/tag/amd"
---

Tag

# #amd

51 posts · 2 followers

[Gayatri Panse](https://hashnode.com/@gradientnotes) · [16h ago](https://hashnode.com/posts/backprop-01-where-did-the-memory-go/6afd92054f515efa31a3329a)

### [Backprop 01: Where Did the Memory Go?](https://gradientdecent.hashnode.dev/backprop-01-where-did-the-memory-go?utm_source=hashnode&utm_medium=feed)

Why the same Gemma 4 12B fit 1.83 users on vLLM and about 13 on llama.cpp

24 min read · [gradientdecent.hashnode.dev](https://gradientdecent.hashnode.dev/backprop-01-where-did-the-memory-go?utm_source=hashnode&utm_medium=feed)

[Cyber Knight](https://hashnode.com/@cyberknight2479) · [2d ago](https://hashnode.com/posts/mior-and-skyfall-the-attacks-that-broke-amd-s-even-the-host-can-t-look-inside-promise/6afaacf01d8bcbf6ec220fc1)

### [Mior and Skyfall: The Attacks That Broke AMD's "Even the Host Can't Look Inside" Promise](https://nulltrace.hashnode.dev/mior-and-skyfall-the-attacks-that-broke-amd-s-even-the-host-can-t-look-inside-promise?utm_source=hashnode&utm_medium=feed)

Two research teams showed a malicious cloud host can read a confidential VM's memory — and forge the signed report that says everything is fine.

5 min read · [nulltrace.hashnode.dev](https://nulltrace.hashnode.dev/mior-and-skyfall-the-attacks-that-broke-amd-s-even-the-host-can-t-look-inside-promise?utm_source=hashnode&utm_medium=feed)

[Gayatri Panse](https://hashnode.com/@gradientnotes) · [Oct 31](https://hashnode.com/posts/epoch-2-it-serves-does-it-train/6ae5a81fd7f0b3f843c8bae8)

### [Epoch 2: It Serves. Does It Train?](https://gradientdecent.hashnode.dev/epoch-2-it-serves-does-it-train?utm_source=hashnode&utm_medium=feed)

Fine-tuning Gemma 4 with Unsloth on two AMD Radeon RX 7900 XTX cards, and how much of the pain was actually AMD's Slug

34 min read · [gradientdecent.hashnode.dev](https://gradientdecent.hashnode.dev/epoch-2-it-serves-does-it-train?utm_source=hashnode&utm_medium=feed)

[Shraddha bhat](https://hashnode.com/@shraddhabhat1909) · [Oct 17](https://hashnode.com/posts/the-8-2-billion-bet-on-the-third-dimension-how-amd-just-changed-the-ai-race-forever/6ad3828b48d1e840ed1438b1)

### [The $8.2 Billion Bet on the Third Dimension: How AMD Just Changed the AI Race Forever](https://aiinsights1909.hashnode.dev/the-8-2-billion-bet-on-the-third-dimension-how-amd-just-changed-the-ai-race-forever?utm_source=hashnode&utm_medium=feed)

The AI Industry Has Been Missing a Dimension. AMD Just Paid $8.2 Billion to Add It.

9 min read · [aiinsights1909.hashnode.dev](https://aiinsights1909.hashnode.dev/the-8-2-billion-bet-on-the-third-dimension-how-amd-just-changed-the-ai-race-forever?utm_source=hashnode&utm_medium=feed)

[Servers99](https://hashnode.com/@servers99) · [Oct 16](https://hashnode.com/posts/knod-explained-offloading-linux-network-traffic-to-amd-gpus/6ad1add04227897b494d8926)

### [KNOD Explained: Offloading Linux Network Traffic to AMD GPUs](https://servers99.hashnode.dev/knod-explained-offloading-linux-network-traffic-to-amd-gpus?utm_source=hashnode&utm_medium=feed)

How the In-Kernel Network Offload Device bypasses CPU bottlenecks and processes up to 70 million packets per second without userspace overhead.

8 min read · [servers99.hashnode.dev](https://servers99.hashnode.dev/knod-explained-offloading-linux-network-traffic-to-amd-gpus?utm_source=hashnode&utm_medium=feed)

[Gayatri Panse](https://hashnode.com/@gradientnotes) · [Oct 13](https://hashnode.com/posts/epoch-1-try-it-time-it-then-buy-it/6ace36b10c270ecc322ef88d)

### [Epoch 1: Try It, Time It, Then Buy It](https://gradientdecent.hashnode.dev/epoch-1-try-it-time-it-then-buy-it?utm_source=hashnode&utm_medium=feed)

How three rounds of Gemma 4 experiments shaped our choice of GPU, inference engine and model

23 min read · [gradientdecent.hashnode.dev](https://gradientdecent.hashnode.dev/epoch-1-try-it-time-it-then-buy-it?utm_source=hashnode&utm_medium=feed)

[MIG servers](https://hashnode.com/@migservers) · [Oct 8](https://hashnode.com/posts/amd-epyc-9006-venice-cpus-architecting-the-agentic-ai-stack/6ac72f8c5083de7953d3216b)

### [AMD EPYC 9006 "Venice" CPUs: Architecting the Agentic AI Stack](https://migservers.hashnode.dev/amd-epyc-9006-venice-cpus-architecting-the-agentic-ai-stack?utm_source=hashnode&utm_medium=feed)

How AMD EPYC 9006 Processors Power Scalable, High-Performance Infrastructure for Agentic AI Workloads

5 min read · [migservers.hashnode.dev](https://migservers.hashnode.dev/amd-epyc-9006-venice-cpus-architecting-the-agentic-ai-stack?utm_source=hashnode&utm_medium=feed)

[Throttle](https://hashnode.com/@throttle-pro) · [Oct 6](https://hashnode.com/posts/the-fp8-trap-my-gpu-bill-dropped-47-because-the-model-was-printing/6ac47d40745f09a5728dcc7f)

### [The FP8 trap: my GPU bill dropped 47% because the model was printing "!!!!!!"](https://throttle.hashnode.dev/the-fp8-trap-my-gpu-bill-dropped-47-because-the-model-was-printing?utm_source=hashnode&utm_medium=feed)

FP8 on an AMD MI300X looked 41-47% cheaper per token. Then I read the answers. What really happened, and the fix that's a real 32% saving.

4 min read · [throttle.hashnode.dev](https://throttle.hashnode.dev/the-fp8-trap-my-gpu-bill-dropped-47-because-the-model-was-printing?utm_source=hashnode&utm_medium=feed)

[Vedant Jadhav](https://hashnode.com/@vedantjadhav) · [Oct 3](https://hashnode.com/posts/amd-llm-lab/6ac15c1df8437aace82ad975)

### [What 57 LLM Benchmark Runs on an AMD Instinct MI300X Taught Me About VRAM, Throughput, and Quantization](https://vedantjadhav.hashnode.dev/amd-llm-lab?utm_source=hashnode&utm_medium=feed)

I built AMD LLM Lab to understand LLM inference performance before running the model.

7 min read · [vedantjadhav.hashnode.dev](https://vedantjadhav.hashnode.dev/amd-llm-lab?utm_source=hashnode&utm_medium=feed)

[3](https://hashnode.com/posts/amd-llm-lab/6ac15c1df8437aace82ad975)

[Josh Green](https://hashnode.com/@joshgreen) · [Aug 1](https://hashnode.com/posts/a-30b-model-crawled-on-my-mini-pc-and-the-weights-were-innocent/6a6db832239173d439c6f5a8)

### [A 30B Model Crawled on My Mini PC, and the Weights Were Innocent](https://joshgreen.hashnode.dev/a-30b-model-crawled-on-my-mini-pc-and-the-weights-were-innocent?utm_source=hashnode&utm_medium=feed)

A mixture-of-experts model fell off a cliff on an AMD mini PC. The fix was a different runtime, not a different model.

5 min read · [joshgreen.hashnode.dev](https://joshgreen.hashnode.dev/a-30b-model-crawled-on-my-mini-pc-and-the-weights-were-innocent?utm_source=hashnode&utm_medium=feed)

[1](https://hashnode.com/posts/a-30b-model-crawled-on-my-mini-pc-and-the-weights-were-innocent/6a6db832239173d439c6f5a8)
