Unverified50% confidenceFactExact time
A conversation with an LLM is essentially an array of alternating user and model messages; when using the API, you must maintain the history yourself, otherwise the model cannot remember past conversations.
1
Sources
50%
Confidence
Long-term
Relevance
10/3/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedThe 'memory' of an LLM is technically implemented by concatenating historical conversation text into the current request's Prompt, allowing the model to see the historical content within a single inference pass.79% similarUnverifiedThe sole reason LLMs can continue a conversation is that the application layer re-feeds the historical context to the model each time.74% similarVerifiedLarge models do not have memory capability by default; a ChatMemory mechanism must be introduced to remember the content of past conversation turns.73% similarUnverifiedThe current mainstream LLM applications adopt a stateless interaction model, where the model itself does not retain cross-session memory and all context must be temporarily fed in via prompts.73% similarUnverifiedMulti-turn dialogue is implemented by re-inputting all previous conversation history to the model with each request, rather than the model having genuine memory capability.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/970242API
curl https://kongchang.com/api/v1/knowledge/claims/970242MCP
get_claim(id=970242)