Hackernews posts about MI200

1. CES 2026: Taking the Lids Off AMD's Venice and MI400 SoCs (chipsandcheese.com)
   135 points by rbanffy 12 days ago | 82 comments
2. AMD Gives Us a Peek at Upcoming Helios System, MI500 GPUs at CES (www.hpcwire.com)
   1 points by rbanffy 11 days ago | discuss
3. AMD's MI300X Outperforms Nvidia's H100 for LLM Inference (www.blog.tensorwave.com)
   280 points by fvv over 1 year ago | 264 comments
4. Testing AMD's Giant MI300X (chipsandcheese.com)
   213 points by klelatti over 1 year ago | 244 comments
5. The AMD Radeon Instinct MI300A's Giant Memory Subsystem (chipsandcheese.com)
   212 points by pella about 1 year ago | 99 comments
6. Porting HPC Applications to AMD Instinct MI300A Using Unified Memory and OpenMP (arxiv.org)
   95 points by arcanus over 1 year ago | 27 comments
7. AMD MI300x GPUs with GEMM tuning improves throughput and latency by up to 7.2x (www.nscale.com)
   66 points by latchkey over 1 year ago | 100 comments
8. Exploring inference memory saturation effect: H100 vs. MI300x (dstack.ai)
   57 points by latchkey about 1 year ago | 12 comments
9. Boosting Computational Fluid Dynamics Performance with AMD MI300X (rocm.blogs.amd.com)
   47 points by latchkey about 1 year ago | 45 comments
10. An EPYC Exclusive for Azure: AMD's MI300C – By George Cozma (chipsandcheese.com)
    41 points by rbanffy about 1 year ago | 14 comments
11. Attention is NOT all you need: Qwerky-72B trained using only 8 AMD MI300X GPUs (substack.recursal.ai)
    20 points by jtatarchuk 10 months ago | 3 comments
12. Linux 6.9 Adding AMD MI300 Row Retirement Support for Problematic HBM Memory (www.phoronix.com)
    20 points by rbanffy almost 2 years ago | 2 comments
13. How the "Antares" MI300 GPU Ramp Will Save AMD's Datacenter Business (www.nextplatform.com)
    19 points by rbanffy almost 2 years ago | discuss
14. AMD MI300x GPUs with GEMM tuning improves throughput and latency by up to 7.2x (www.nscale.com)
    16 points by yankcrime over 1 year ago | discuss
15. AMD MI300X vs. Nvidia H100 LLM Benchmarks (blog.runpod.io)
    13 points by aadillpickle over 1 year ago | 4 comments
16. Early LLM serving experience and performance results with AMD Instinct MI300X (blogs.oracle.com)
    12 points by googlechromer over 1 year ago | 1 comments
17. Benchmarking Llama 3.1 405B on 8x AMD MI300X GPUs (dstack.ai)
    11 points by latchkey over 1 year ago | 3 comments
18. Harnessing AI Compute Power Atop Open-Source Software: 8 X AMD MI300X (www.phoronix.com)
    10 points by rbanffy almost 2 years ago | discuss
19. Using AMD MI300X for High-Throughput, Low-Cost LLM Inference (www.herdora.com)
    8 points by technoabsurdist 6 months ago | discuss
20. AMD's Long and Winding Road to the Hybrid CPU-GPU Instinct MI300A (www.nextplatform.com)
    6 points by rbanffy over 1 year ago | discuss
21. MI300X vs. H100 vs. H200 Benchmark Part 1: Training (newsletter.semianalysis.com)
    5 points by nl about 1 month ago | discuss
22. AMD MI300X performance compared with Nvidia H100 (www.tomshardware.com)
    5 points by latchkey over 1 year ago | discuss
23. Nvidia H100 vs. AMD MI300X (blog.runpod.io)
    4 points by shaamilkarim over 1 year ago | 1 comments
24. Inter-APU Communication on AMD MI300A Systems via Infinity Fabric: A Deep Dive (arxiv.org)
    4 points by matt_d 5 months ago | discuss
25. DeepSeek R1 inference performance: MI300X vs. H200 (dstack.ai)
    4 points by latchkey 10 months ago | discuss
26. Sizing Up MI300A's GPU (chipsandcheese.com)
    4 points by klelatti 12 months ago | discuss
27. How the "Antares" MI300 GPU Ramp Will Save AMD's Datacenter Business (www.nextplatform.com)
    4 points by rbanffy almost 2 years ago | discuss
28. Take AMD MI300X for a test drive (tensorwave.com)
    3 points by jtatarchuk almost 2 years ago | 3 comments
29. Microsoft beat H200 Deepseek inference with MI300 (techcommunity.microsoft.com)
    3 points by galeos 9 months ago | 1 comments
30. HPE Launches AMD EPYC Venice Instinct MI400 and Nvidia Vera Rubin Compute Blades (www.servethehome.com)
    3 points by PaulHoule about 2 months ago | discuss
31. My First Multi-GPU Kernel: Writing All-to-All for AMD MI300X (gau-nernst.github.io)
    3 points by mfiguiere 3 months ago | discuss
32. Dissecting CPU-GPU Unified Physical Memory on AMD MI300A APUs (arxiv.org)
    3 points by matt_d 5 months ago | discuss