Qwen3.8-Flash-Next
Alibaba (Qwen)MoE experts: 512 experts, 10 routed + 1 shared activated per token
coremodel architecturein Qwen3.8-Flash-NextQwen Team
implementation detail · filed under model architecture
A configuration activating 10 routed experts and 1 shared expert for each token.
The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See shared experts for the methods filed beside it.
Also called 10 routed + 1 shared activated per token, 10 Routed + 1 Shared.
2 quotes from 2 sources, one card per model, each opening with its strongest.
MoE experts: 512 experts, 10 routed + 1 shared activated per token
Number of Activated Experts: 10 Routed + 1 Shared