Verified65% confidenceFactExact time
The thinking budget mechanism of reasoning models sets an upper limit on the number of reasoning tokens, and once exhausted, the model cannot output a complete answer, potentially causing overthinking issues.
3
Sources
65%
Confidence
Long-term
Relevance
7/9/2026
First Seen
Sources
16GB VRAM Showdown: Ornith 35B vs Qwen 35B — Full Benchmark Comparison
bilibili程序员-智能译站7/7/2026
Related Claims
UnverifiedSome failure cases can be traced to the model continuing to reason and exhausting the budget that should have been used to give the final answer, ultimately failing to converge to a conclusion.78% similarUnverifiedChain-of-thought reasoning models may generate hundreds or even thousands of thinking tokens even for extremely simple problems, causing latency to spike and inference costs to surge.78% similarVerifiedModels may fall into the trap of overthinking, as longer reasoning time does not always lead to better results.77% similarUnverifiedUsers can balance quality and cost by setting the thinking budget parameter to cap the reasoning token limit, rather than unconditionally using the highest reasoning intensity.76% similarUnverifiedIn evaluation, if the thinking budget is not fixed, scores of the same model are not comparable, and the budget effectively becomes a hidden hyperparameter.76% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/342899API
curl https://kongchang.com/api/v1/knowledge/claims/342899MCP
get_claim(id=342899)