What is the point of MoE models, beyond being faster?

<!-- SC_OFF --><div class="md"><p>is it possible that a moe model can decide how many billion parameters to activate per token according to the task. eg if qwen 3.6 35b a3b - if a

<!-- SC_OFF --><div class="md"><p>Found this a few months back for more compute-efficient (relative to memory size) MoE models. Nowadays Qwen3.6-35B-A3B exists along with

<table> <tr><td> <a href="https://www.reddit.com/r/LocalLLaMA/comments/1s59rct/rys_part_3_llms_think_in_geometry_not_language/"> <img alt="RYS Part 3: LLMs think in geometry, not

<!-- SC_OFF --><div class="md"><p>&#x200b;</p> <p>## Abstract</p> <p>Post-training makes language models more decisive without necessarily making them more accurate — and we find

<!-- SC_OFF --><div class="md"><p>Forgive me naiveness, I’m a little lost. Why must a model be jack of all trades? Why can’t we have a model expert in a single thing?</p>

<!-- SC_OFF --><div class="md"><p>Long time Claude Opus user, but after the recent session limit changes by Anthropic, I am seriously considering trying Chinese models for coding.

<!-- SC_OFF --><div class="md"><p><strong>Noam Brown</strong> — OpenAI Research Scientist and co-creator of o1, Libratus, and Cicero — sat down with Dwarkesh Patel for a 1h 20m

<!-- SC_OFF --><div class="md"><p>Hey, I’ve been really excited to see the latest models being released, but I keep wondering: what are we actually supposed to do with them?</p>