specific method · filed under model architecture
DeepSeekMoE
The named MoE framework identified as retained in the cited DeepSeek models.
Also called DeepSeekMoE framework.
- sources
- 2
- models
- 2
- lab adopt it
- 1
- strongest
- core
How sources treat it
One count per evidence span, weakest treatment to strongest.
used 1core 1
Documented in
Evidence
2 spans quoted from the sources, strongest treatment first.
DeepSeek-V4 series retain the DeepSeekMoE framework
coremodel architecturein DeepSeek-V4DeepSeek
retains the DeepSeekMoE framework and Multi-Token Prediction (MTP) strategy from DeepSeek-V3
usedmodel architecturein DeepSeek-V4DeepSeek
Filed alongside
Other methods under model architecture :: channel mixer :: mixture of experts.
Mixture of ExpertsSparse expert activationGated DeltaNet MoEMixture-of-Experts layersMoE with routed and shared expertsAsymmetric input/output activation splitDispatch recomputationFrequency-based partitioning of N-gram embedding slotsHybrid Mixture of ExpertsMegaMoEMoE with 128 routed experts and a shared expertMoE with 256 experts and top-8 routingMoE with 384 routed experts and a shared expertOutput-independent MoE gradient reformulationRouted expert output modulationSigmoid-based MoE router with auxiliary-loss-free load balancing