Unverified50% confidenceDecision RuleExact time
The calculation method for rates quoted per million tokens is: divide each usage category by one million first, multiply by the corresponding rate, then add the results together.
1
Sources
50%
Confidence
Long-term
Relevance
10/14/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAssuming each call consumes 2,000 tokens at a combined input/output price of $0.002, the inference cost for one million calls reaches $2,000.75% similarUnverifiedCalculated at the paired public unit prices ($0.28 per million input, $1.10 per million output), the 400-case single hosted run costs $1.1061, and τ=0.4 costs $0.0747.74% similarUnverifiedThe inference cost of AI models is primarily determined jointly by token consumption volume and unit price, typically priced per one million tokens.72% similarUnverifiedLLM pricing is typically billed per million tokens, with input and output often using different rates.72% similarUnverifiedOne API 的额度系统默认以 1:500000 的比例将美元换算为内部额度单位，实现跨模型的成本核算71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1030186API
curl https://kongchang.com/api/v1/knowledge/claims/1030186MCP
get_claim(id=1030186)