Unverified80% confidenceOpinionExact time
For models that are called repeatedly, inference costs already dominate the total lifecycle expenditure of a model, and when N×c_infer×T is much greater than C_train, the total cost of ownership is dominated by inference.
1
Sources
80%
Confidence
Long-term
Relevance
8/19/2026
First Seen
Sources
Rethinking Scaling Laws: Parameters Are Not the Only Answer
twitterjietang8/19/2026
Related Entities
Related Claims
UnverifiedThe total inference cost of large models exceeds the training cost, because inference continues to occur and grows linearly with user scale.78% similarUnverifiedThe internal chain of thought in reasoning models generates a large number of output tokens, making the actual cost of completing complex tasks often far higher than what a per-token price comparison would suggest.75% similarUnverifiedThe inference cost of large language models is variable and usage-based; a single Agent handling a complex task may trigger dozens of model calls and consume millions of tokens.73% similarUnverifiedReasoning models (such as the o1 series) can cost several times to more than ten times more than regular models because they need to generate a large number of intermediate thinking steps.73% similarUnverifiedExtended thinking mode causes the model to perform an explicit internal reasoning chain before output, at the cost of significantly increased token consumption, with reasoning tokens typically billed at a higher rate.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/796928API
curl https://kongchang.com/api/v1/knowledge/claims/796928MCP
get_claim(id=796928)