Unverified50% confidenceEventExact time
In a call with about ninety-seven thousand-plus input tokens, only the newest one thousand eight hundred-plus tokens were not cached; without caching the input cost was about 0.1 yuan, and with caching it was less than 0.004 yuan, 26 times cheaper.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
11/6/2026
First Seen
Valid until: 11/20/2026
Sources
How Token Caching Can Save You 98% on Your DeepSeek Bill
bilibili雪舞☆血雾10/22/2026
Related Entities
Related Claims
UnverifiedA calculation example: asking 20 questions about a 50,000-token document requires about 1 million input tokens without prompt caching, and about 157,000 with caching, reducing input cost by about 84%, with faster time to first token as well.79% similarUnverifiedIn real-world testing of 1.78 million token long conversations, the configuration that sent the most tokens was actually the cheapest because 97% of the tokens were cached.77% similarUnverifiedIn one long conversation of 36 turns with approximately 1.78 million tokens, the full-retention configuration was actually the cheapest because 97% of the tokens hit the cache.74% similarUnverifiedMemo is free when saving fewer than ten thousand times and searching fewer than one thousand times per month, and the cost of running a local model can be zero.73% similarUnverifiedThe actual extent of cache savings depends on the hit rate; if 80% of input hits the cache, the actual input cost is approximately 0.8 × cache price + 0.2 × regular price.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1142187API
curl https://kongchang.com/api/v1/knowledge/claims/1142187MCP
get_claim(id=1142187)