Unverified50% confidenceFactExact time
Reasoning tokens are no longer double-counted, but cost calculations themselves are not affected.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/4/2026
First Seen
Valid until: 1/2/2027
Sources
Related Entities
Related Claims
UnverifiedCost accounting should separately track the volume and cost of input tokens and output tokens, rather than looking at total token count in aggregate.72% similarUnverifiedThinking tokens for reasoning models are typically billed, and setting a token budget cap can prevent runaway costs from a single request.72% similarUnverifiedThe token multiplier scheme has controllable transformation costs for existing subscription systems, with core changes only in the usage statistics module and frontend display, requiring no modification to the underlying inference architecture.68% similarUnverifiedReasoning content is often billed the same as regular text in terms of token counting, so retaining reasoning increases the number of tokens carried in requests and affects call costs.66% similarVerifiedIn API billing, the unit price of output tokens is generally higher than that of input tokens, because the generation process requires autoregressive inference token by token64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/971390API
curl https://kongchang.com/api/v1/knowledge/claims/971390MCP
get_claim(id=971390)