Reverie: An Inference-Time Cognitive Architecture for Continuous LLM Deliberation

LLMs are reactive: a prompt arrives, the model computes, and cognition halts until the next prompt. Existing multi-agent LLM systems inherit this constraint: they coordinate multiple calls per response but still respond only when prompted. We introduce Reverie, an inference-time architecture in which a set of specialised agents runs continuously on a shared workspace where ideas accumulate, decay, and compete for influence. An executive composes a response when the user speaks, or on its own when accumulated internal activity crosses a threshold; between turns, the system keeps thinking. On a downstream task of research assistance over $60$ open-ended scientific questions, five independent human raters, blinded to system identity, preferred Reverie's conversations over Mixture-of-Agents on $201/300$ pairs ($67\%$; mixed-effects $95\%$ CI $0.62$–$0.72$, $p=1.7\times 10^{-7}$), with every rater preferring Reverie. At matched compute, Reverie also produces $16\%$ more conceptually distinct thoughts per turn than Mixture-of-Agents and $106\%$ more than Self-Consistency (Vendi Score, Bonferroni-significant), at roughly $3\times$ lower cloud-equivalent cost per session than Mixture-of Agents. On nine verifiable benchmarks (factual QA, math, code) Reverie is statistically tied with Mixture-of-Agents on eight and worse on rare-fact retrieval (PopQA), so the cost of continuous deliberation on convergent tasks is small.