Qwen2.5-1.7T-A70B

Qwen2.5-1.7T-A70B is a sparse mixture-of-experts language model from Alibaba, featuring 1.7 trillion total parameters with 70 billion active per token. It delivers high performance at reduced inference cost, ranking second only to Anthropic's Claude

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.