Hackernews posts about MI300X

1. 01
   

   newVibeSys Builds a Qwen3.5-397B Engine on MI300A, 2.3× Tuned SGLangsyfi.cs.washington.edu

   ▲ 2matt_d7ddiscusssavesaved
2. 02
   

   newDeepSeek V4 Flash on a Single AMD MI300Xgithub.com

   ▲ 382zhoutong74d💬 109savesaved
3. 03
   

   newBringing Up DeepSeek-V4-Flash on AMD MI300Xfergusfinn.com

   ▲ 120kkm137d💬 25savesaved
4. 04
   

   newExploring inference memory saturation effect: H100 vs. MI300xdstack.ai

   ▲ 57latchkey681d💬 12savesaved
5. 05
   

   newBoosting Computational Fluid Dynamics Performance with AMD MI300Xrocm.blogs.amd.com

   ▲ 47latchkey639d💬 45savesaved
6. 06
   

   newWhy isn't AMD's MI300X competitive?newsletter.semianalysis.com

   ▲ 46colonCapitalDee172d💬 36savesaved
7. 07
   

   newAttention is NOT all you need: Qwerky-72B trained using only 8 AMD MI300X GPUssubstack.recursal.ai

   ▲ 20jtatarchuk564d💬 3savesaved
8. 08
   

   newCross-vendor byte-identical inference for a 72B LLM (AMD MI300X vs. Nvidia H100)zenodo.org

   ▲ 9ashsyngh53ddiscusssavesaved
9. 09
   

   newUsing AMD MI300X for High-Throughput, Low-Cost LLM Inferencewww.herdora.com

   ▲ 8technoabsurdist462ddiscusssavesaved
10. 10
    

    newMI300X vs. H100 vs. H200 Benchmark Part 1: Trainingnewsletter.semianalysis.com

    ▲ 5nl314ddiscusssavesaved
11. 11
    

    newDeepSeek R1 inference performance: MI300X vs. H200dstack.ai

    ▲ 4latchkey578ddiscusssavesaved
12. 12
    

    newA Fast Attention Kernel for MI300X, Written in Hip, Not Assemblymoonmath.ai

    ▲ 3latchkey118ddiscusssavesaved
13. 13
    

    newRepomind – 256K context coding agent on a single AMD MI300X (FP8)github.com

    ▲ 3sardor_r1162ddiscusssavesaved
14. 14
    

    newMy First Multi-GPU Kernel: Writing All-to-All for AMD MI300Xgau-nernst.github.io

    ▲ 3mfiguiere348ddiscusssavesaved
15. 15
    

    newReal world(ish) DeepSeek V4 Flash performance on a single MI300Xmatthusby.github.io

    ▲ 2mhusby57ddiscusssavesaved
16. 16
    

    newShow HN: Chisel – Profile AMD MI300X kernels locallygithub.com

    ▲ 2technoabsurdist482ddiscusssavesaved
17. 17
    

    newBenchmarking MI300X Memcpyscalarlm.ghost.io

    ▲ 2latchkey534ddiscusssavesaved
18. 18
    

    newSupercharge DeepSeek-R1 Inference on AMD Instinct MI300Xrocm.blogs.amd.com

    ▲ 2pama574ddiscusssavesaved
19. 19
    

    newUnlock DeepSeek-R1 Inference Performance on AMD Instinct MI300X GPUrocm.blogs.amd.com

    ▲ 2breadislove601ddiscusssavesaved
20. 20
    

    newVLLM X AMD: Efficient LLM Inference on AMD Instinct MI300X GPUs (Part 1)www.amd.com

    ▲ 2latchkey643ddiscusssavesaved
21. 21
    

    newAMD Data Center GPUs Explained: MI250X, MI300X, MI350X and Beyondwww.bentoml.com

    ▲ 1djhu9408ddiscusssavesaved
22. 22
    

    newAMD MI300X for LLM Serving Disaggregating Prefill and Decode with SGLangrocm.blogs.amd.com

    ▲ 1latchkey414ddiscusssavesaved
23. 23
    

    newAMD MI300X Memcpy Peer Deep Divewww.scalarlm.com

    ▲ 1gdiamos523ddiscusssavesaved
24. 24
    

    newCranking Out Faster Tokens for Fewer Dollars: AMD MI300X vs. Nvidia H200eliovp.com

    ▲ 1latchkey532ddiscusssavesaved
25. 25
    

    newMI300X vs. H100 vs. H200 Benchmarksemianalysis.com

    ▲ 1chongzhe663ddiscusssavesaved
26. 26
    

    newBreaking CUDA Boundaries: Hashcat Runs Natively on Hot Aisle AMD MI300x Scalewww.youtube.com

    ▲ 1latchkey690ddiscusssavesaved
27. 27
    

    newStory of How Im Running an Unlimited $6/Month AI Provider on 4x RTX 3090s

    ▲ 8yolo-auto125d💬 3savesaved
28. 28
    

    newAsk HN: Is anyone using AMD GPUs for their AI workloads?

    ▲ 6technoabsurdist479d💬 2savesaved
29. 29
    

    newAsk HN: Workstation/rig config for local LLM finetuning/training and inference?

    ▲ 3behnamoh549d💬 1savesaved
30. 30
    

    newShow HN: CLI tool for discovering AMD compute resources from multiple providersgithub.com

    ▲ 3cjavelona474ddiscusssavesaved
31. 31
    

    newShow HN: Chisel – Profile GPU Kernels Without a GPU (Nvidia and AMD)github.com

    ▲ 3technoabsurdist477ddiscusssavesaved

· edit