MoE language model with 32B active and 1T total parameters.
Cross-referenced across 55 tracked directories
#315
Popularity Rank
1 / 55
Listed In
Emerging
Adoption Stage
3d
Listed For
Recently added to the ecosystem
Fully open reproduction of DeepSeek-R1
Pushing the frontier of cost-effective reasoning.
Clean, minimal, accessible reproduction of DeepSeek R1-Zero
Exploring the Intelligence of Large-scale MoE Model.