Expired50% confidenceBenchmarkExact time
After the compression layer was introduced in the previous version, the token consumption for a page-open action dropped from approximately 100,000 to 40,000
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/16/2026
First Seen
Valid until: 10/14/2026(expired)
Sources
Related Claims
UnverifiedSome users have reported generating 450 million or even 3.5 billion input tokens in a single day, with actual costs reduced by dozens of times at a 99% cache hit rate63% similarUnverifiedByteDance Ark's subscription starts at 40 yuan per month and supports the latest Doubao Side 2.0 model.62% similarVerifiedBy 2025, efficient models such as GPT-4o mini have compressed the cost per million tokens to less than $0.2, a reduction of more than 99%.59% similarUnverifiedUsing Prompt Caching technology, the same API call costs $1 the first time and only $0.05 the second time, reducing costs by 20 times.59% similarUnverifiedA $50 price reduction in the $300–$400 price segment represents a price cut of approximately 12%–15%.58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/533617API
curl https://kongchang.com/api/v1/knowledge/claims/533617MCP
get_claim(id=533617)