Recalld, a memory layer for AI agents.
Raw Developer Origin & Technical Request
Hacker News
Sep 30, 2026
Recalld is a memory layer for AI agents and chats. It can ingest conversations, documents or code. The input is decomposed into facts, then saved into a vector database.
I built Recalld because I couldn't find a reliable memory layer that let my agents retrieve accurate data without noise.The facts from each add operation are compared with existing facts, and Recalld then decides whether each one is an addition, an update or a replacement of existing facts.
Agents can retrieve context via two methods. The first is search, which is faster and cheaper and performs a vector search.
The second option is recall which uses an LLM to select relevant facts. In our published benchmark on LoCoMo benchmark run, it returned an average of about 243 context tokens per question. See the benchmarks for more details: github.com/bit-robotics/reca... can be accessed via API or MCP. Benchmarks are public and can be rerun.Recalld has a free tier plan with no card required. We offer two completely separate regions, EU and US, depending on where you want your data to be stored.Recalld is hosted only and cannot be self-hosted yet
Developer Debate & Comments
No active discussions extracted for this entry yet.
Frequently Asked Questions
Market intelligence mapped to Recalld, a memory layer for AI agents..
What problem does Recalld, a memory layer for AI agents. solve?
What architecture is tied to Recalld, a memory layer for AI agents.?
Engagement Signals
Cross-Market Term Frequency
Counts stored text matches for foundational terms like memory layer for AI agents and chats and ingest conversations, documents or code in analysed public discussions. These counts do not measure adoption.
Stored text matches, not adoption or market-share measurements.
- memory layer for AI agents and chats: 1 records
- ingest conversations, documents or code: 1 records
- decomposed into facts: 1 records
- vector database: 1 records
- retrieve accurate data without noise: 1 records
- add operation: 1 records
- addition, update or replacement of existing facts: 1 records
- retrieve context: 1 records
- search (vector search): 1 records
- recall (LLM to select relevant facts): 1 records
- LoCoMo benchmark: 1 records
- 243 context tokens per question: 1 records
- API or MCP: 1 records
- public benchmarks: 1 records
- free tier plan: 1 records
- EU and US regions: 1 records
- hosted only: 1 records
SaaS Metrics