Long Claude Code sessions can get costly after a 1-hour break
Long Claude Code can use much more of a Fable allowance when they are paused for more than an hour and then continued. Claude needs to read the earlier conversation when handling a new request, so a long chat carries a large context. keeps that context on the server and lets later requests read it at about 10% of the normal input price.
When the cache expires or is in, the next request has to write the whole context again, which costs much more. For Fable 5, normal input is $10 per million tokens, a cache read is $1 per million tokens, a cache write is $12.50 per million tokens in the 5-minute window, and $20 per million tokens in the 1-hour cache window. In a heavy session with 400,000 tokens of context, a normal cached turn costs about $0.40, but returning after more than an hour can create about $8 of .
One expired cache can cost about as much as 20 normal cached turns.
Key points
- Long Claude Code become more expensive as the saved context grows.
- makes repeated use of a long conversation much cheaper while the cache is still active.
- After more than an hour, the cache may expire and force a full cache write.
- With 400,000 tokens of context, one expired cache is estimated at about $8 of .
- A single cache expiry can cost about the same as 20 normal cached turns.