A desktop computer allegedly trains a 2.8T-parameter MoE model with just one active expert
max_paperclips · x · 2026-07-21
A repost jokes about “training a 2.8T parameter model” on a desktop computer, adding that only one expert is active and it is always the same one. The humor comes from the absurdity of describing a giant MoE model as if it were running on a normal desktop.
- It is clearly a joke about sparse MoE routing.
- The punchline is the fake simplicity of “one expert active.”
- This is the kind of AI in-joke that spreads well on social platforms.
More from Fun
- $ANSEM airdrop sends one user $2,114 and sparks a crypto payoff joke — mfckr_eth · 2026-07-21
- Kimi K3 builds a 3D football stadium and a fuller samurai scene in one user showcase — eyishazyer · 2026-07-21
- Kimi K3 turns a samurai prompt into a fuller scene and powers a procedural planet inspector — eyishazyer · 2026-07-21
- A rebuilt Windows XP runs Vice City and Red Alert 2, then circles back to Moonshot’s VR buddy — eyishazyer · 2026-07-21
- A planet inspector and a rebuilt Windows XP round out the Kimi K3 demo thread — eyishazyer · 2026-07-21
- Moonshot’s Kimi VR companion can listen, reply, show expressions, and move across scenes — eyishazyer · 2026-07-21