Hackernews posts about HB100

  1. Show HN: ExANS – Lossless KV cache compression at 622 GB/s on H100 (www.theopenlake.com)
    15 points by arnav__1 16 days ago | discuss
  2. LFM2.5-DSpark: Up to 3.2x Faster Inference from H100 to MacB (www.liquid.ai)
    13 points by Alephinitesimal about 7 hours ago | discuss
  3. Proving Muse Glimmer in Zero-Knowledge at 85 Tok/S on an H100 (attestable.com)
    4 points by mildog8 10 days ago | discuss
  4. See how an H100 works in 3D with threejs (gpu.kylejeong.com)
    3 points by Kylejeong21 6 days ago | discuss
  5. vLLM Serving Experiments on H100s – config beats the baseline on p95 TTFT,ITL (efficientagent.substack.com)
    1 points by bnayak25 12 days ago | discuss
  6. Show HN: Distilling DeepSeek into GPT-OSS doesn't transfer censorship. Try it (www.ctgt.ai)
    170 points by cgorlla 22 days ago | 73 comments
  7. Launch HN: Discovered Materials (YC P26) – AI agents to discover new materials (discoveredmaterials.com)
    162 points by advaith08 10 days ago | 36 comments
  8. Launch HN: machine0 (YC S26) – Persistent CPU and GPU VMs from the CLI (machine0.io)
    81 points by bwm 3 days ago | 44 comments
  9. Launch HN: Stoa Markets (YC S26) – A Marketplace for GPUs and AI Servers (www.stoaexchange.com)
    77 points by erenberke 11 days ago | 51 comments
  10. Show HN: Cuts Long Horizon Inference Costs by 50% via external KV Cache Offload (github.com)
    22 points by arnav__1 26 days ago | discuss
  11. Show HN: I built a hypervisor and client for inference on consumer compute (scalattice.com)
    3 points by RomulusHill 27 days ago | discuss
  12. Ask HN: You have model orders of magnitudes faster than transformer, so?
    1 points by aunekiaru 6 days ago | discuss
  13. Casio F-B100W-1A (www.casio.com)
    462 points by __fst__ 2 days ago | 384 comments
  14. $2 H100s: How the GPU Rental Bubble Burst (www.latent.space)
    403 points by swyx almost 2 years ago | 279 comments
  15. Deploying DeepSeek on 96 H100 GPUs (lmsys.org)
    285 points by GabrielBianconi 12 months ago | 80 comments
  16. Processing Strings 109x Faster Than Nvidia on H100 (ashvardanian.com)
    216 points by ashvardanian 11 months ago | 26 comments
  17. Furiosa: 3.5x efficiency over H100s (furiosa.ai)
    212 points by written-beyond 7 months ago | 160 comments
  18. Huawei's Ascend 910C delivers 60% of Nvidia H100 inference performance (www.tomshardware.com)
    118 points by sien over 1 year ago | 61 comments
  19. Exploring inference memory saturation effect: H100 vs. MI300x (dstack.ai)
    57 points by latchkey over 1 year ago | 12 comments
  20. Colossus AI Supercluster with over 100k Nvidia H100 GPUs (twitter.com)
    47 points by DeathArrow almost 2 years ago | 33 comments
  21. Processing Strings 109x Faster Than Nvidia on H100 (ashvardanian.com)
    34 points by samspenc 11 months ago | 3 comments
  22. How to train a model on 10k H100 GPUs? (soumith.ch)
    33 points by bobrenjc93 almost 2 years ago | discuss
  23. Dual RTX 5090 Beats $25,000 H100 in Real-World LLM Performance (www.hardware-corner.net)
    29 points by mdp2021 over 1 year ago | 12 comments
  24. Rent H100 SXMs for 0.99 (app.hyperbolic.xyz)
    10 points by hyperbolic_labs almost 2 years ago | 5 comments
  25. Ask HN: What is the cheapest ($/hour) H100 renting service with quick spin-up?
    10 points by dtquad almost 2 years ago | 4 comments
  26. Meta is using more than 100k Nvidia H100 AI GPUs to train Llama-4 (www.tomshardware.com)
    8 points by turtlegrids almost 2 years ago | 1 comments
  27. xAI's Colossus (100k H100 cluster) has begun training (twitter.com)
    7 points by famouswaffles almost 2 years ago | 1 comments
  28. H100 prices are melting *UP (www.latent.space)
    7 points by swyx 5 months ago | discuss
  29. Processing Strings 109x Faster Than Nvidia on H100 (ashvardanian.com)
    6 points by binarymax 11 months ago | discuss
  30. Nvidia B200 vs. H100: Independent Training Performance Tests (www.lightly.ai)
    6 points by isusmelj over 1 year ago | discuss
  31. MI300X vs. H100 vs. H200 Benchmark Part 1: Training (newsletter.semianalysis.com)
    5 points by nl 9 months ago | discuss
  32. Show HN: A GPU group-buying layer to get enterprise rates (~50% off H100s) (trysaga.ai)
    5 points by DdotR over 1 year ago | discuss
  33. Cohere Open-Sources Command A+, a 218B Moe Model That Runs on Two H100s (firethering.com)
    4 points by steveharing1 3 months ago | discuss
  34. Outperforming cuBLAS on H100: A Worklog (cudaforfun.substack.com)
    4 points by skidrow over 1 year ago | discuss
  35. Elon Musk's xAI Colossus supercomputer with 100K H100 Nvidia GPUs comes online (www.tomshardware.com)
    4 points by Link_ke almost 2 years ago | discuss
  36. The "setup tax" on AWS H100s is killing iterative research
    3 points by miyamotomusashi 8 months ago | 2 comments
  37. Tracing tokens through Llama 3.1 8B inference on H100s (krithik.xyz)
    3 points by krithik_7 3 months ago | 1 comments
  38. A100s are more cost-efficient than H100s for small training tasks (cloudbreakr.substack.com)
    3 points by jack_nclx 8 months ago | 1 comments
  39. The Great GPU Shortage: H100 Rental Prices Up 40% (newsletter.semianalysis.com)
    3 points by alecco 4 months ago | discuss
  40. AI Subscription vs. H100 [video] (www.youtube.com)
    3 points by owenpalmer 7 months ago | discuss
  41. H100 PCIe – 1.86 TB/s memcpy roofline and 8× uplift
    3 points by GPUrouter 11 months ago | discuss
  42. Outperforming cuBLAS on H100: A Worklog (cudaforfun.substack.com)
    3 points by skidrow over 1 year ago | discuss
  43. Outperforming cuBLAS on H100: A Worklog (cudaforfun.substack.com)
    3 points by skidrow over 1 year ago | discuss