Hackernews posts about MI300

1. The AMD Radeon Instinct MI300A's Giant Memory Subsystem (chipsandcheese.com)
   212 points by pella 12 days ago | 99 comments
2. Boosting Computational Fluid Dynamics Performance with AMD MI300X (rocm.blogs.amd.com)
   47 points by latchkey 15 days ago | 45 comments
3. Sizing Up MI300A's GPU (chipsandcheese.com)
   4 points by klelatti 10 days ago | discuss
4. Sizing Up MI300A's GPU – By Chester Lam (chipsandcheese.com)
   2 points by rbanffy 9 days ago | discuss
5. VLLM X AMD: Efficient LLM Inference on AMD Instinct MI300X GPUs (Part 1) (www.amd.com)
   2 points by latchkey 18 days ago | discuss
6. Sizing up MI300A's GPU (chipsandcheese.com)
   1 points by rbanffy 7 days ago | discuss
7. Sizing Up MI300A's GPU – By Chester Lam (chipsandcheese.com)
   1 points by rbanffy 8 days ago | discuss
8. The AMD Radeon Instinct MI300A's Giant Memory Subsystem (old.chipsandcheese.com)
   1 points by ksec 12 days ago | discuss
9. AMD's MI300X Outperforms Nvidia's H100 for LLM Inference (www.blog.tensorwave.com)
   280 points by fvv 8 months ago | 264 comments
10. Testing AMD's Giant MI300X (chipsandcheese.com)
    213 points by klelatti 7 months ago | 244 comments
11. AMD MI300X 30% higher performance than Nvidia H100, even with optimized stack (www.tomshardware.com)
    169 points by g42gregory about 1 year ago | 94 comments
12. AMD MI300 performance – Faster than H100, but how much? (www.semianalysis.com)
    116 points by klelatti about 1 year ago | 85 comments
13. AMD Expands AI Product Lineup with GPU-Only Instinct Mi300X with 192GB Memory (www.anandtech.com)
    114 points by mfiguiere over 1 year ago | 104 comments
14. Azure announces new AI optimized VM series featuring AMD's flagship MI300X GPU (techcommunity.microsoft.com)
    95 points by latchkey about 1 year ago | 67 comments
15. Porting HPC Applications to AMD Instinct MI300A Using Unified Memory and OpenMP (arxiv.org)
    95 points by arcanus 9 months ago | 27 comments
16. AMD Instinct MI300 Data Center APU – 146B Transistors, Shipping H2’23 (www.anandtech.com)
    70 points by kristianp almost 2 years ago | 40 comments
17. AMD's Response to Nvidia regarding H100 vs. MI300X on real world workloads (community.amd.com)
    70 points by treesciencebot about 1 year ago | 16 comments
18. AMD MI300x GPUs with GEMM tuning improves throughput and latency by up to 7.2x (www.nscale.com)
    66 points by latchkey 7 months ago | 100 comments
19. AMD AI Software Solved – MI300X Pricing, Perf, PyTorch, FlashAttention, Triton (www.semianalysis.com)
    60 points by another over 1 year ago | 23 comments
20. Exploring inference memory saturation effect: H100 vs. MI300x (dstack.ai)
    57 points by latchkey about 2 months ago | 12 comments
21. AMD Launches Instinct MI300X AI Accelerator, Up to 60% Faster Than Nvidia H100 (wccftech.com)
    48 points by fariszr about 1 year ago | 16 comments
22. An EPYC Exclusive for Azure: AMD's MI300C – By George Cozma (chipsandcheese.com)
    41 points by rbanffy about 2 months ago | 14 comments
23. AMD Instinct MI300X Accelerators (www.amd.com)
    36 points by bx376 about 1 year ago | 27 comments
24. Linux 6.9 Adding AMD MI300 Row Retirement Support for Problematic HBM Memory (www.phoronix.com)
    20 points by rbanffy 12 months ago | 2 comments
25. AMD ROCm 6.0 Now Available with MI300 Support, PyTorch FP8 and More AI (www.phoronix.com)
    19 points by latchkey about 1 year ago | 1 comments
26. How the "Antares" MI300 GPU Ramp Will Save AMD's Datacenter Business (www.nextplatform.com)
    19 points by rbanffy 12 months ago | discuss
27. AMD MI300x GPUs with GEMM tuning improves throughput and latency by up to 7.2x (www.nscale.com)
    16 points by yankcrime 7 months ago | discuss
28. AMD MI300X vs. Nvidia H100 LLM Benchmarks (blog.runpod.io)
    13 points by aadillpickle 7 months ago | 4 comments
29. Early LLM serving experience and performance results with AMD Instinct MI300X (blogs.oracle.com)
    12 points by googlechromer 7 months ago | 1 comments
30. Benchmarking Llama 3.1 405B on 8x AMD MI300X GPUs (dstack.ai)
    11 points by latchkey 4 months ago | 3 comments
31. Harnessing AI Compute Power Atop Open-Source Software: 8 X AMD MI300X (www.phoronix.com)
    10 points by rbanffy 11 months ago | discuss
32. AMD can get ~65% more mi300x from the same 5nm wafer than Nvidia can get H100s (old.reddit.com)
    9 points by latchkey about 1 year ago | 17 comments
33. AMD MI300 Ramp, GPT-4 Performance, Asp and Volumes (www.semianalysis.com)
    8 points by rbanffy about 1 year ago | 1 comments
34. Giagantic AMD APU at SC23 Meet the AMD MI300A and MI300X (www.servethehome.com)
    8 points by rbanffy about 1 year ago | discuss
35. AMD Unleashes MI300X AI GPU to Rival Nvidia's Market Supremacy (www.news.upveda.in)
    8 points by mysweetmisery over 1 year ago | discuss
36. AMD MI300 AI Performance, Volume Ramp, Customers, Cost, IO, Networking, Software (www.semianalysis.com)
    7 points by rbanffy over 1 year ago | discuss
37. AMD's Long and Winding Road to the Hybrid CPU-GPU Instinct MI300A (www.nextplatform.com)
    6 points by rbanffy 7 months ago | discuss
38. AMD MI300: AI Performance, Volumes, Customers, Cost, IO, Networking, Software (www.semianalysis.com)
    5 points by rbanffy over 1 year ago | 2 comments