DeepSeek-V3 671B

DeepSeek-V3 671B is a large Mixture-of-Experts (MoE) AI model developed by DeepSeek, known for its massive 671 billion total parameters and efficient training, achieving record speeds of approximately two minutes on CoreWeave's MLPerf v6.0 benchmarks

Two-hop subgraph: this entity, every entity it directly relates to, and every entity those neighbors relate to. Drag a node, scroll to zoom, click to inspect — or click any neighbor and re-center the atlas there.