Living Brain pitches one shared memory for ChatGPT, Claude, Cursor, Codex and your own agents
The maker claims 92.2% on LoCoMo and 464ms median retrieval.
Christel Buchanan (@ladyxtel) introduced Living Brain, which she says decouples memory from the model: one persistent memory you write to in one AI tool and retrieve from in any other.
The post claims 92.2% on the LoCoMo benchmark, 464ms median retrieval and up to 22.6x fewer tokens per question. These are the maker's own numbers.
It drew about 1,000 likes.
