Memrith is a desktop application that sends prompts and selected portions of your data to the AI you have configured — either a third-party AI service (such as Anthropic, OpenAI, or OpenRouter) using an API key you provide, or a local model you run on your own machine (such as Ollama, llama.cpp, or LM Studio). Every response, summary, claim, suggestion, analysis, recommendation, or other piece of content that appears in Memrith is generated by that AI model — not by Memrith LLC.
Memrith does not produce, fact-check, validate, or endorse the AI's output. Memrith makes no representation or warranty that any output is accurate, complete, current, appropriate, safe, lawful, free of bias, or fit for any particular purpose. Output may be:
Memrith also extracts memories from this output and from your entries; those memories can inherit the same errors. You can review, correct, or delete any memory at any time.
Memrith does not control how you use what the AI generates. You are responsible for:
You must not use Memrith — or rely on its output — as a substitute for professional advice in:
without independent review by a qualified human professional.
To the fullest extent permitted by law, Memrith LLC disclaims all liability for any harm, loss, error, damage, or legal exposure arising from output or from your use of it — including but not limited to: financial loss, reputational harm, legal liability, medical or health consequences, business decisions, personal relationships, and any reliance on output that turns out to be incorrect, biased, hallucinated, or otherwise mistaken.
This disclaimer is in addition to (not instead of) the warranty disclaimers and the limitation of liability in the Terms of Service. Where you use a third-party AI Provider, your contract for those AI services runs directly between you and that provider; Memrith is not a party to it. If you run a local model yourself, no third-party provider is involved.
Contact us at legal@memrith.com.