3 ms·

(2025) As it's 9 months old and they just had a major model release

• GaggiX 3mo ago
  I believe OP posted it because the new Kimi K3 has 69 KDA layers (the rest are 24 Gated MLA), I think previous large Kimi models had only MLA layers.
  • yorwba 3mo ago
    It's not the same KDA as used in Kimi Linear, though.
    • GaggiX 3mo ago
      What's the difference? They are both called Kimi Delta Attention.
      • yorwba 3mo ago
        The differences are explained in section 2.1.1 of the Kimi K3 technical report: https://arxiv.org/pdf/2607.24653#page=4 https://arxiv.org/pdf/2607.24653#page=4
• throwa356262 3mo ago
  For K3 read this instead: https://arxiv.org/abs/2607.24653 https://arxiv.org/abs/2607.24653 The main contribution of the K3 paper is Stable LatentMoE. Like some other models it compresses data sent between layers, which puts certain requirements on the router. K3 improves performance by using a more balanced expert selection strategy.
  • mcbuilder 3mo ago
    Compared to the Opus 5 "model card", which read like a standard Anthropic set of alignment principles and safety concerns, this presents a plethora of useful technical details that advances the state of the art.
    • throwa356262 3mo ago
      Same with DeepSeek papers, they are a joy to read.
  • senko 3mo ago
    Not an expert, but looks like they did a lot more work on the RL part (9 expert models, full sandbox access for agentic tasks, etc)?
    • verdverm 3mo ago
      most new effort in training comes in the late phase with RL techniques the pretraining (slurping the internet) only goes so far, the new data being used is from human preferences and agent traces (designed and/or distilled)
• cptcobalt 3mo ago
  Rather under-discussed back then: https://news.ycombinator.com/item?id=45766937 https://news.ycombinator.com/item?id=45766937