Hackernews posts about PyTorch

1. 01
   

   newGoogle open-sources TorchTPU, a native PyTorch back end for TPUsgithub.com

   ▲ 380ilreb16d💬 93savesaved
2. 02
   

   new0/1 Knapsack Problem in PyTorchjedrzej.maczan.pl

   ▲ 52yu3zhou422d💬 7savesaved
3. 03
   

   newTorchTPU is now open source: Run native PyTorch on TPUscloud.google.com

   ▲ 8wondersky16d💬 1savesaved
4. 04
   

   newReplacing PyTorch with Julia and Reactant: One Year Incsvance.dev

   ▲ 7csvance14d💬 1savesaved
5. 05
   

   newPyTorch-native TPU back end for vLLMgithub.com

   ▲ 6zniturah17ddiscusssavesaved
6. 06
   

   newHow to Run Turing-Complete Malware in PyTorch's Backward Passelonlit.com

   ▲ 5czhou57815d💬 1savesaved
7. 07
   

   newAmazon opens its AWS Trainium AI chips to every PyTorch developerwww.aboutamazon.com

   ▲ 4axolotl_raja16ddiscusssavesaved
8. 08
   

   newTorchNeuron: Native PyTorch Back End for AWS Trainiumpytorch.org

   ▲ 4wondersky17ddiscusssavesaved
9. 09
   

   newGPUlayer: Train PyTorch on every GPU you can reach-faster than PyTorch Lightningwww.gpulayer.dev

   ▲ 3antonellof4d💬 1savesaved
10. 10
    

    newHow to Run Turing-Complete Malware in PyTorch's Backward Passelonlit.com

    ▲ 3simonpure14ddiscusssavesaved
11. 11
    

    newReplacing PyTorch with Julia and Reactant: One Year Incsvance.dev

    ▲ 3leephillips14ddiscusssavesaved
12. 12
    

    newShow HN: Tensor Golf – Practise PyTorch Through Short Drillstensor.golf

    ▲ 3singular_atomic20ddiscusssavesaved
13. 13
    

    newAI Development on Windows: From PyTorch and Llama.cpp to Windows MLdevblogs.microsoft.com

    ▲ 3antimora29ddiscusssavesaved
14. 14
    

    newOvermind: Cut PyTorch model load time 15s to 0.2s with a zero-copy shared cachegithub.com

    ▲ 2westurner17d💬 1savesaved
15. 15
    

    newPyTorch 2.15 Release Blogpytorch.org

    ▲ 2wondersky7ddiscusssavesaved
16. 16
    

    newTorchNeuron: Native PyTorch Back End for AWS Trainiumpytorch.org

    ▲ 2gmays15ddiscusssavesaved
17. 17
    

    newRun your PyTorch models on encrypted dataeml.belfortlabs.com

    ▲ 2furkanturan16ddiscusssavesaved
18. 18
    

    newTorchTPU: Hybrid Eager/Compiled Architecture for Native PyTorch Execution on TPUarxiv.org

    ▲ 1matt_d14d💬 1savesaved
19. 19
    

    newBlog FlexAttention: The Flexibility of PyTorch with the Performance of FApytorch.org

    ▲ 1skidrow7ddiscusssavesaved
20. 20
    

    newShow HN: Macvoice – A local voice agent for Mac in 1.4 GB, no PyTorchgithub.com

    ▲ 1johnrobinsn7ddiscusssavesaved
21. 21
    

    newShow HN: Simulusion – A PyTorch library for training and running world modelsgithub.com

    ▲ 1paramthakkar9ddiscusssavesaved
22. 22
    

    newShow HN: Mini-batch k-means clustering for PyTorch tensorsgithub.com

    ▲ 1mb1412ddiscusssavesaved
23. 23
    

    newThe State of PyTorch 2026techstrong.ai

    ▲ 1CrankyBear14ddiscusssavesaved
24. 24
    

    newAmazon's TorchNeuron Runs PyTorch on AWS Trainium Chipswww.aboutamazon.com

    ▲ 1jonbaer15ddiscusssavesaved
25. 25
    

    newMPs in PyTorch 2.15huggingface.co

    ▲ 1subset16ddiscusssavesaved
26. 26
    

    newShow HN: TensorViz – Explore PyTorch models inside VS Codetensorviz.dev

    ▲ 1nityanandmathur25ddiscusssavesaved
27. 27
    

    newShow HN: Open-source alternative to AWS BlazingText, in PyTorchgithub.com

    ▲ 1Fazel9427ddiscusssavesaved
28. 28
    

    newAI Development on Windows: From PyTorch and Llama.cpp to Windows MLdevblogs.microsoft.com

    ▲ 1pjmlp28ddiscusssavesaved
29. 29
    

    newShow HN: Who and where are your anagranames?louisabraham.github.io

    ▲ 3Labo33332hdiscusssavesaved
30. 30
    

    newShow HN: Sweet-search – local code search for agents (5–33% cheaper runs)github.com

    ▲ 3mrsladoje4ddiscusssavesaved

· edit