Hackernews posts about HB100

  1. Dual RTX 5090 Beats $25,000 H100 in Real-World LLM Performance (www.hardware-corner.net)
    29 points by mdp2021 19 days ago | 12 comments
  2. Nvidia B200 vs. H100: Independent Training Performance Tests (www.lightly.ai)
    6 points by isusmelj 10 days ago | discuss
  3. Outperforming cuBLAS on H100: A Worklog (cudaforfun.substack.com)
    4 points by skidrow 27 days ago | discuss
  4. Outperforming cuBLAS on H100: A Worklog (cudaforfun.substack.com)
    3 points by skidrow 19 days ago | discuss
  5. Outperforming cuBLAS on H100: A Worklog (cudaforfun.substack.com)
    1 points by skidrow 26 days ago | discuss
  6. Ask HN: Where to Resell Cloud GPUs?
    2 points by mertleee 9 days ago | 3 comments
  7. HB100 Doppler Radar Module Teardown (www.allaboutcircuits.com)
    67 points by metadat 10 months ago | 5 comments
  8. Show HN: San Francisco Compute – 512 H100s at <$2/hr for research and startups (sfcompute.org)
    727 points by flaque over 1 year ago | 176 comments
  9. $2 H100s: How the GPU Rental Bubble Burst (www.latent.space)
    403 points by swyx 6 months ago | 279 comments
  10. So you want to rent an NVIDIA H100 cluster? 2024 Consumer Guide (www.photoroom.com)
    297 points by ea016 10 months ago | 139 comments
  11. AMD's MI300X Outperforms Nvidia's H100 for LLM Inference (www.blog.tensorwave.com)
    280 points by fvv 10 months ago | 264 comments
  12. Nvidia H100 GPUs: Supply and Demand (gpus.llm-utils.org)
    227 points by tin7in over 1 year ago | 160 comments
  13. Google Launches AI Supercomputer Powered by Nvidia H100 GPUs (www.tomshardware.com)
    194 points by jonbaer almost 2 years ago | 176 comments
  14. Nvidia H100 and A100 GPUs – comparing available capacity at GPU cloud providers (llm-utils.org)
    183 points by tikkun almost 2 years ago | 126 comments
  15. Google TPU v5p beats Nvidia H100 (www.techradar.com)
    174 points by wslh about 1 year ago | 133 comments
  16. AMD MI300X 30% higher performance than Nvidia H100, even with optimized stack (www.tomshardware.com)
    169 points by g42gregory over 1 year ago | 94 comments
  17. H100 GPUs Set Standard for Gen AI in Debut MLPerf Benchmark (blogs.nvidia.com)
    157 points by Anon84 almost 2 years ago | 75 comments
  18. Intel Gaudi2 chips outperform Nvidia H100 on diffusion transformers (stability.ai)
    146 points by memossy about 1 year ago | 60 comments
  19. Nvidia’s H100: Funny L2, and Tons of Bandwidth (chipsandcheese.com)
    132 points by picture almost 2 years ago | 49 comments
  20. Huawei's Ascend 910C delivers 60% of Nvidia H100 inference performance (www.tomshardware.com)
    118 points by sien 2 months ago | 61 comments
  21. Cloud Provider Gets $2.3B Loan Using Nvidia's H100 as Collateral (www.anandtech.com)
    117 points by belltaco over 1 year ago | 44 comments
  22. AMD MI300 performance – Faster than H100, but how much? (www.semianalysis.com)
    116 points by klelatti over 1 year ago | 85 comments
  23. Tesla turns on 10k-node Nvidia H100 Cluster (www.techradar.com)
    80 points by Kevcmk over 1 year ago | 129 comments
  24. IBM analog AI chip could give the Nvidia H100 a run for its money (www.techradar.com)
    78 points by mikece over 1 year ago | 60 comments
  25. AMD's Response to Nvidia regarding H100 vs. MI300X on real world workloads (community.amd.com)
    70 points by treesciencebot over 1 year ago | 16 comments
  26. NVIDIA introduces TensorRT-LLM for accelerating LLM inference on H100/A100 GPUs (developer.nvidia.com)
    69 points by mkaushik over 1 year ago | 21 comments
  27. U.S. Bans Sales of Nvidia's H100, A100 GPUs to Middle East (www.tomshardware.com)
    57 points by ironyman over 1 year ago | 68 comments
  28. Exploring inference memory saturation effect: H100 vs. MI300x (dstack.ai)
    57 points by latchkey 5 months ago | 12 comments
  29. Nvidia L40S is a Nvidia H100 AI alternative (www.servethehome.com)
    51 points by PaulHoule over 1 year ago | 27 comments
  30. AMD Launches Instinct MI300X AI Accelerator, Up to 60% Faster Than Nvidia H100 (wccftech.com)
    48 points by fariszr over 1 year ago | 16 comments
  31. Colossus AI Supercluster with over 100k Nvidia H100 GPUs (twitter.com)
    47 points by DeathArrow 6 months ago | 33 comments
  32. AWS P5 instances with H100 generally available (aws.amazon.com)
    35 points by rushingcreek over 1 year ago | 3 comments
  33. How to train a model on 10k H100 GPUs? (soumith.ch)
    33 points by bobrenjc93 7 months ago | discuss
  34. Intel's "Gaudi 3" AI accelerator chip may give Nvidia's H100 a run for its money (arstechnica.com)
    32 points by justinclift about 1 year ago | 10 comments
  35. Bay Bridge: the cheapest H100 training clusters (sfcompute.com)
    30 points by flaque over 1 year ago | 8 comments
  36. Nvidia H100 GPU Shipments by Customer (www.threads.net)
    28 points by ahiknsr over 1 year ago | 24 comments