Unverified50% confidenceSolutionExact time
The correct prompt budget formula is: budget = window - reserved_output - schemas - system, meaning reserved output tokens, tool schemas, and the system prompt must all be subtracted from the window first.
1
Sources
50%
Confidence
Long-term
Relevance
10/19/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedSetting a tool-call budget for each task and returning a structured budget_exceeded prompt when the budget is exhausted can switch the model from continuing to try to summarizing the information it has obtained.73% similarVerifiedThe context window refers to the upper limit of input tokens a model can accommodate in a single inference; the system prompt, user messages, tool results and conversation history share this budget, and once exceeded, it is either truncated or an error is returned.66% similarUnverifiedThe context window is not a warehouse, but a budget table that must be re-orchestrated every turn.66% similarUnverifiedToken budget refers to the maximum token quota allocated to different context components (system prompt, retrieved documents, conversation history, tool outputs, etc.) within a single request or a single session.66% similarUnverified任务编排循环通常由框架代码而非模型本身控制，设定最大迭代次数、超时限制、成本预算等边界条件63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1071674API
curl https://kongchang.com/api/v1/knowledge/claims/1071674MCP
get_claim(id=1071674)