Reference Architecture · Agents
CoALA: how language agents remember, decide and act
A reference architecture for thinking about any LLM agent: three kinds of long-term memory, a working memory where the model reasons, an action space, and a decision cycle that ties them together. Explained with two diagrams and one worked example.
Agent frameworks use the same words for different things. "Memory" might mean the chat history, a vector database or a fine-tuned model. "Planning" might be one prompt or a search over many. Without shared vocabulary, two designs are hard to compare and gaps are hard to spot.