OpenAI’s New Model Shows Why Caching Matters
AI Wallet
OpenAI has launched GPT-6.1 Sol with the same standard API price as GPT-6 Sol: $2 per million input tokens and $10 per million output tokens. But cached input costs only $0.10 per million tokens, 95% less than normal input. Caching allows AI applications to reuse information they have already processed instead of paying the full input price every time.
Illustrative example assuming the first request costs $2 and subsequent requests qualify for the $0.10 cached-input rate. Actual savings depend on usage and cache eligibility.
This can make a big difference for businesses using AI agents, long instructions or the same documents repeatedly. The advertised model price has not fallen, but the cost of reusing information can be much lower. This shows why comparing AI models only by their standard token price does not tell the full story. How a model is used can increasingly matter as much as which model is selected.