λ-RLM

λ-RLM is an 8 billion parameter recursive language model from MIT that processes arbitrarily long contexts by recursively summarizing its own outputs, enabling efficient long-context reasoning.

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.