memtrust provides an independent, reproducible benchmark harness for evaluating agent-memory backends. It is designed for AI researchers and developers who need to assess and compare the performance of memory systems in agent architectures. The tool is open source and runs as a CLI utility.
In the LLM eval & observability space, memtrust takes a focused approach. It enables developers to independently and reproducibly benchmark agent-memory backend systems. It is built as an open-source project for AI researchers and developers. memtrust is open source under the Apache-2.0 license. It runs on the command line.
memtrust first shipped in 2026. Key capabilities include benchmark harness, reproducible evaluation, and agent memory testing.
Latest indexed changes and source events
Other apps tracked under the same category.
