Smart News

Hackernews posts about MoE

RSS feed for MoE
  1. 61
    Strata: Qwen3.8-Flash-Next (125B Moe) on a 8GB+ Nvidia GPUgithub.com
    ▲ 5simonpure6ddiscuss
  2. 62
    StepFun teases Step 5 Preview, a 600B/27B MoE, open weights coming Oct 15twitter.com
    ▲ 3WavyPeng13d💬 1
  3. 63
    SSD-Llama: SSD-Native Inference for Trillion-Parameter Moe on a Consumer PCarxiv.org
    ▲ 3AlmostCosmo7916ddiscuss
  4. 64
    Show HN: Llama.cpp fork with 2-4x multiGPU speed for MoE models bigger than VRAMgithub.com
    ▲ 2neuralll9d💬 3
  5. 65
    I joined the 'anti-chatbox' interface movement, and build Moe because of itmoebot.app
    ▲ 2yisrael_g19d💬 1
  6. 66
    Show HN: Moe Routing Atlas – which experts fire in Qwen3.5/3.6 35B-A3Bgslaller.github.io
    ▲ 2gslaller5ddiscuss
  7. 67
    Xing4.0-29B-A4B: China Telecom's Moe Model Trained on Ascend NPUswww.mindstudio.ai
    ▲ 2Bluestein14ddiscuss
  8. 68
    Show HN: I made a tiny MoE/Engram viz toolsw-ml-study.github.io
    ▲ 2softwarewright21ddiscuss
  9. 69
    The Newsroom EP01 – OpenAI Ships GPT-5 to Azure – DeepSeek Open-Sources 236B Moewww.youtube.com
    ▲ 2LebToki23ddiscuss
  10. 70
    Empirical Research on MoE Safety Mechanismsdealign.ai
    ▲ 2Bluestein24ddiscuss
  11. 71
    Extending Raschka's GPT-2: an MoE trained from scratch on an RTX 3090www.gilesthomas.com
    ▲ 2ibobev24ddiscuss
  12. 72
    Dense vs. Moe Models: Active Parameters, Throughput, and When to Choose Eachdeveloper.nvidia.com
    ▲ 1pretext19ddiscuss
  13. 73
    Extending Raschka's GPT-2: an MoE trained from scratch on an RTX 3090www.gilesthomas.com
    ▲ 1gpjt24ddiscuss
  14. 74
    Fractal-BLT – Zero-allocation .NET 10 MoE runtime streaming NVMe to GPUgithub.com
    ▲ 1Ctrl_Alt_Haze27ddiscuss
  15. 75
    Show HN: Mini-AGI – Dynamic continual learning model trained on 8GB VRAMgithub.com
    ▲ 277volotat14d💬 81
  16. 76
    Ask HN: What do AI improvements at these labs look like?
    ▲ 2xyzsparetimexyz21ddiscuss
  17. 77
    Show HN: Way to divide parameter space in LLM trainingzenodo.org
    ▲ 1genji9708ddiscuss
  18. 78
    Show HN: PicoLM v1.0-rc2 ("Yura Kana"). Run an LLM on Digital Unixgithub.com
    ▲ 1gabucino11ddiscuss
  19. 79
    Reverse engineering my e-scooter and rewriting the firmware in Rustbensimms.moe
    ▲ 452vinhnx25d💬 105
  20. 80
    Don't call yourself an artisanal programmerpurplesyringa.moe
    ▲ 57emschwartz22d💬 65
  21. 81
    Vibe Coding Production Kit – a production workflow for AI coding agentsgithub.com
    ▲ 9Moeeryani10d💬 6
  22. 82
    Grieving the Loss of Detailspurplesyringa.moe
    ▲ 6ibobev10ddiscuss
  23. 83
    Write While Learningpurplesyringa.moe
    ▲ 6dalvrosa15ddiscuss
  24. 84
    Grieving the Loss of Detailspurplesyringa.moe
    ▲ 4ibobev9d💬 2
  25. 85
    Write While Learningpurplesyringa.moe
    ▲ 3meetpateltech16ddiscuss
  26. 86
    Squeezing rust onto more things, this time: Flashlightsbensimms.moe
    ▲ 3Bluestein21ddiscuss
  27. 87
    Evergardenevergarden.moe
    ▲ 3birdculture23ddiscuss
  28. 88
    We Spoke with Entrepreneurs Living the "7 Vibe 3" Lifestylecadence.moe
    ▲ 2xyzsparetimexyz25d💬 3
  29. 89
    Evergardenevergarden.moe
    ▲ 2birdculture13ddiscuss
  30. 90
    Reverse engineering my E-Scooter and rewriting it in Rustbensimms.moe
    ▲ 2nitros26ddiscuss
More stories →

Keyboard shortcuts

j / k
next / previous story
o or Enter
open the story
c
open the comments
s
save for later
h
history & saved stories
?
show this help