Expired50% confidenceFactExact time
The large model FC layer costs approximately $0.002 per request, the vector routing layer costs approximately $0.0001 per request, and the L0 regex layer has zero cost.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
Related Claims
UnverifiedIndustry estimates place the marginal cost of a single complex query to mainstream large models at between $0.01 and $0.1.77% similarUnverifiedThe cost of manual annotation per item is typically between $0.1 and $1, while the marginal cost of model evaluation can be as low as one-thousandth of that74% similarUnverifiedOn an NVIDIA H100 GPU, running a model of approximately 200 billion parameters (Opus scale), the cost of generating 1,000 tokens in a single inference is approximately $0.01–$0.03.71% similarUnverifiedDomestic AI image generation services such as Tongyi Wanxiang and Wenxin Yige have per-image costs ranging from 0.1 to 0.5 RMB.71% similarUnverifiedThe cost of processing a batch of outputs using a top-tier model is approximately $25, whereas a budget model handles the same task for less than $1.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/48901API
curl https://kongchang.com/api/v1/knowledge/claims/48901MCP
get_claim(id=48901)