Unverified50% confidenceFactExact time
Claude 4's extended reasoning can be combined with tool use, and the model is able to reason about when to call functions, interpret results, reason again, and coherently chain actions over long spans.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/15/2026
First Seen
Valid until: 1/13/2027
Sources
Related Entities
Related Claims
VerifiedTool calling allows models to dynamically invoke external functions during inference and incorporate execution results into the context of subsequent reasoning, forming a closed loop of perception, decision-making, and execution.76% similarUnverified推理时计算扩展通过延长思维链、多路径采样择优、自我验证反思等方式在推理阶段投入更多计算可提升复杂推理任务性能74% similarVerifiedClaude's Extended Thinking mode causes the model to perform an explicit internal reasoning chain before output, improving complex task quality but significantly increasing token consumption.73% similarVerifiedWhen Claude's Extended Thinking feature is enabled, the model outputs an internal reasoning process (a thinking block) before generating the final response; this reasoning is visible to the user but is not counted as part of the conversation context.73% similarUnverifiedThe Thinking reasoning phase of Claude Opus 5.5 will always use Adaptive mode, dynamically adjusting reasoning depth according to task complexity.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1035603API
curl https://kongchang.com/api/v1/knowledge/claims/1035603MCP
get_claim(id=1035603)