Hackernews posts about HB100

RSS feed for HB100

01
newShow HN: H200 GPUs under $1.99/hour, H100s at $1.19/hourcompute.cheap
▲ 5EmiCorleone 28d discuss
02
newH200 GPUs $1.99/hour, H100s at $1.19/hourcompute.cheap
▲ 2gpuguy 28d discuss
03
newShow HN: Zero downtime embedding model upgradesgithub.com
▲ 6coolArnav 28d 💬 3
04
newIn Japan, Television Viewing Doesn't Have to Take a Back Seat (2013)www.wsj.com
▲ 3libpcap 25d discuss
05
new$2 H100s: How the GPU Rental Bubble Burstwww.latent.space
▲ 403swyx 725d 💬 279
06
newDeploying DeepSeek on 96 H100 GPUslmsys.org
▲ 285GabrielBianconi 402d 💬 80
07
newProcessing Strings 109x Faster Than Nvidia on H100ashvardanian.com
▲ 216ashvardanian 381d 💬 26
08
newFuriosa: 3.5x efficiency over H100sfuriosa.ai
▲ 212written-beyond 264d 💬 160
09
newHuawei's Ascend 910C delivers 60% of Nvidia H100 inference performancewww.tomshardware.com
▲ 118sien 607d 💬 61
10
newShow HN: The cheapest GPU cloud – H100s at $2.04/HR, H200s at $3/HRcompute.cheap
▲ 63EmiCorleone 33d 💬 20
11
newExploring inference memory saturation effect: H100 vs. MI300xdstack.ai
▲ 57latchkey 669d 💬 12
12
newColossus AI Supercluster with over 100k Nvidia H100 GPUstwitter.com
▲ 47DeathArrow 702d 💬 33
13
newProcessing Strings 109x Faster Than Nvidia on H100ashvardanian.com
▲ 34samspenc 379d 💬 3
14
newDual RTX 5090 Beats $25,000 H100 in Real-World LLM Performancewww.hardware-corner.net
▲ 29mdp2021 552d 💬 12
15
newLFM2.5-DSpark: Up to 3.2x Faster Inference from H100 to MacBwww.liquid.ai
▲ 18Alephinitesimal 45d discuss
16
newShow HN: ExANS – Lossless KV cache compression at 622 GB/s on H100www.theopenlake.com
▲ 15arnav__1 61d discuss
17
newRent H100 SXMs for 0.99app.hyperbolic.xyz
▲ 10hyperbolic_labs 684d 💬 5
18
newAsk HN: What is the cheapest ($/hour) H100 renting service with quick spin-up?
▲ 10dtquad 708d 💬 4
19
newCross-vendor byte-identical inference for a 72B LLM (AMD MI300X vs. Nvidia H100)zenodo.org
▲ 9ashsyngh 41d discuss
20
newMeta is using more than 100k Nvidia H100 AI GPUs to train Llama-4www.tomshardware.com
▲ 8turtlegrids 703d 💬 1
21
newH100 prices are melting *UPwww.latent.space
▲ 7swyx 191d discuss
22
newProcessing Strings 109x Faster Than Nvidia on H100ashvardanian.com
▲ 6binarymax 379d discuss
23
newNvidia B200 vs. H100: Independent Training Performance Testswww.lightly.ai
▲ 6isusmelj 543d discuss
24
newMI300X vs. H100 vs. H200 Benchmark Part 1: Trainingnewsletter.semianalysis.com
▲ 5nl 302d discuss
25
newShow HN: A GPU group-buying layer to get enterprise rates (~50% off H100s)trysaga.ai
▲ 5DdotR 606d discuss
26
newProving Muse Glimmer in Zero-Knowledge at 85 Tok/S on an H100attestable.com
▲ 4mildog8 55d discuss
27
newCohere Open-Sources Command A+, a 218B Moe Model That Runs on Two H100sfirethering.com
▲ 4steveharing1 135d discuss
28
newOutperforming cuBLAS on H100: A Worklogcudaforfun.substack.com
▲ 4skidrow 560d discuss
29
newThe "setup tax" on AWS H100s is killing iterative research
▲ 3miyamotomusashi 278d 💬 2
30
newTracing tokens through Llama 3.1 8B inference on H100skrithik.xyz
▲ 3krithik_7 148d 💬 1
31
newA100s are more cost-efficient than H100s for small training taskscloudbreakr.substack.com
▲ 3jack_nclx 298d 💬 1
32
newShow HN: Compute.cheap – the cheapest H100/H200 compute in the worldcompute.cheap
▲ 3EmiCorleone 34d discuss
33
newSee how an H100 works in 3D with threejsgpu.kylejeong.com
▲ 3Kylejeong21 51d discuss
34
newThe Great GPU Shortage: H100 Rental Prices Up 40%newsletter.semianalysis.com
▲ 3alecco 177d discuss