90% on LoCoMo, 91–93% on LongMemEval

Backboard.io today published results on the two leading AI memory benchmarks, LoCoMo and LongMemEval, reinforcing its position as the foundational AI stack for production-grade and agentic systems.

On the full 500-question LongMemEval, four separate judges scored Backboard's answers between 91.2% and 93.4%. The highest of the four counts 467 of 500 answers as correct.

These results build on Backboard's previously published 90.0% accuracy on the LoCoMo benchmark. Scoring well on both benchmarks is uncommon, as most systems optimize for either short-horizon precision or long-horizon persistence, but not both.

Independent Validation of What "Memory" Really Means

In a recent analysis published by the Ottawa Business Journal, Adyasha Maharana, creator of the LoCoMo benchmark and research scientist at Databricks, clarified an important distinction:

"The system built by Backboard.io is a far better attempt at simulating memory as it manifests in humans. It is practical, cheaper, scalable and doesn't rely solely on brute-force LLM processing for answers."

A Complete AI Stack, Not a Bolt-On Component

Backboard.io is not a router, a wrapper, or a memory plugin. It is a unified AI infrastructure stack. From a single API, Backboard provides:

• Persistent long-term memory

• Native embeddings and vectorization

• Retrieval-augmented generation (RAG)

• Shared memory across agents

• Access to more than 17,000 large language models, including BYOK

Making Agentic AI Practical

Backboard addresses the core constraint of agentic AI by enabling persistent, shared memory across countless agents, even when those agents operate on different underlying models.

"Agentic AI doesn't become meaningful because you call something an agent," said Imbeault. "It becomes meaningful when agents can remember, coordinate, and operate over time. Solving memory is the prerequisite."

Built by a Founder Enterprises Already Trust

Imbeault previously founded Assent, a platform trusted by Fortune 100 companies. "Enterprise systems don't get to reset," said Imbeault. "If they lose context or trust, they fail. That mindset shaped how we built Backboard."

What Comes Next

With foundational memory validated, Backboard is turning its attention to Switchboard, a new capability designed to help developers and enterprises better understand how different AI system configurations behave under real-world constraints.