Remembrall
Long-term memory for large language models
Subscription3,757 impressions#180 of its week
Remembrall adds long-term memory and document retrieval to OpenAI chat calls via a proxy.
- for
- Developers building LLM-powered agents or apps that need persistent user context.
- pricing
- freemium
- license
- AGPL-3.0
Key features
- User-ID memory — Stores facts from a conversation and re-injects the five closest memories on the next call.
- Document retrieval — Uploads document sets and adds relevant passages to prompts via vector search.
- Short-term history — Replays the last 20 messages as short-term context with the x-gp-short header.
- Dashboard observability — Free UI logs all LLM requests for monitoring and debugging.
- OpenAI proxy compatibility — Works as a drop-in replacement for the OpenAI Chat Completions endpoint.
- Simple integration — Add two headers (x-gp-remember, x-gp-context) and a base URL to enable memory.
Use cases
- Personalised chatbot that remembers a user's past interactions.
- QA assistant that pulls information from uploaded knowledge bases.
- Customer support agent that retains ticket history across sessions.
- Internal tool that logs and reuses facts extracted from conversations.
Remembrall FAQ
Can I retrieve my old Remembrall memories or logs?+
No, the service and its database went offline in August 2024 and the data is not recoverable.
Is the current remembrall.dev site the original service?+
No, the domain was re-registered in October 2026 and is not affiliated with the original team.
Is the Remembrall source code available?+
Yes, the code is on GitHub (reductoai/remembrall) under AGPL-3.0, but it is unmaintained and depends on retired components.
Should I rotate my OpenAI API key used with Remembrall?+
Yes, because the key was passed through the proxy and the service is no longer operational.
How do I replace Remembrall in a LangChain app?+
Remove the custom base URL and x-gp-* headers, then use LangGraph's Store for long-term memory with the same user ID namespace.
Summarized by DevHunt from remembrall.dev · Oct 4, 2026. Details may change; check the official site.
