Unverified50% confidenceFactExact time
When making decisions, an agent generates plans and possesses planning and memory capabilities, whereas in Function Calling the model inference only decides whether to call a function, with no planning or memory by default.
1
Sources
50%
Confidence
Long-term
Relevance
9/17/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedFunction Calling采用关注点分离设计，语言模型只负责决策，真正的工具执行由应用层代码负责，模型不直接触碰外部系统73% similarUnverifiedIn an agent, the large model is responsible for reasoning and planning decisions, while tools are responsible for executing tasks, with tool invocation uniformly controlled and decided by the large model.73% similarUnverifiedFunction Calling是模型层面的原子能力，而一个Skill可能包含多次Function Calling的编排、错误处理、重试策略和结果后处理71% similarUnverified智能体的执行具有非确定性，同样的输入可能因模型的随机采样而产生不同输出，因此其恢复机制需要保存完整的执行轨迹和中间推理步骤71% similarUnverifiedReasoning configuration allows developers to control the model's reasoning effort, determining how much reasoning computation the model invests before answering.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/930515API
curl https://kongchang.com/api/v1/knowledge/claims/930515MCP
get_claim(id=930515)