Unverified50% confidenceFactExact time
The Memory node in n8n uses Window Buffer Memory by default.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/5/2026
First Seen
Valid until: 1/3/2027
Sources
Related Entities
Related Claims
Unverifiedn8n supports multiple memory backends, including windowed buffer short-term memory, vector database-based long-term memory (such as Pinecone and Supabase Vector), and summary-based compressed memory.78% similarVerifiedAn agent's memory includes short-term memory (conversational context window, typically 4K–128K tokens), long-term memory (vector database), and procedural memory (intermediate state cache for tool calls).61% similarUnverifiedShort-term working memory, i.e., the context window, is typically 8K to 128K tokens, while long-term memory is usually stored in the form of a vector database61% similarUnverifiedThe core objective of Zero-Mem is to achieve zero-token memory operations—that is, to provide LLM agents with persistent memory read/write capabilities without consuming any of the model's context window tokens.60% similarUnverifiedmmBERT supports an input window of up to 8192 tokens.58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/977172API
curl https://kongchang.com/api/v1/knowledge/claims/977172MCP
get_claim(id=977172)