Prompt caching pricing compared
89 models charge less for input they have seen before. When the start of your prompt repeats across requests, such as a long system prompt or a document you ask several questions about, those tokens cost typically 90% less.
Prices updated October 28, 2026
Cached input prices per million tokens
| Model | Provider | Input / 1M | Cached input / 1M | Discount |
|---|---|---|---|---|
| deepseek-flash | DeepSeek | $0.3 | $0.006 | 98% |
| deepseek-v4-flash | DeepSeek | $0.3 | $0.006 | 98% |
| deepseek-v4-1-flash-1017 | Mistral | $0.3 | $0.007 | 98% |
| claude-fable-5-1 | Anthropic | $10 | $0.25 | 98% |
| claude-mythos-5-1 | Anthropic | $10 | $0.25 | 98% |
| deepseek-v4-pro | DeepSeek | $1.32 | $0.044 | 97% |
| gpt-6.1-terra | OpenAI | $1 | $0.05 | 95% |
| gpt-6.1-sol | OpenAI | $2 | $0.1 | 95% |
| claude-sonnet-5-5 | Anthropic | $2 | $0.1 | 95% |
| gemini-4-argon | Google Gemini | $2 | $0.1 | 95% |
| claude-opus-5-5 | Anthropic | $4 | $0.2 | 95% |
| gpt-5-nano | OpenAI | $0.05 | $0.005 | 90% |
| gpt-6-luna | OpenAI | $0.1 | $0.01 | 90% |
| claude-haiku-5-5 | Anthropic | $0.1 | $0.01 | 90% |
| gemini-2.5-flash-lite | Google Gemini | $0.1 | $0.01 | 90% |
| ministral-3-3b-2512 | Mistral | $0.1 | $0.01 | 90% |
| ministral-3b-2512 | Mistral | $0.1 | $0.01 | 90% |
| mistral-small | Mistral | $0.1 | $0.01 | 90% |
| deepseek-coder | DeepSeek | $0.14 | $0.014 | 90% |
| ministral-3-8b-2512 | Mistral | $0.15 | $0.015 | 90% |
| ministral-8b-2512 | Mistral | $0.15 | $0.015 | 90% |
| gpt-5.4-nano | OpenAI | $0.2 | $0.02 | 90% |
| gpt-5.6-luna | OpenAI | $0.2 | $0.02 | 90% |
| ministral-3-14b-2512 | Mistral | $0.2 | $0.02 | 90% |
| ministral-14b-2512 | Mistral | $0.2 | $0.02 | 90% |
| gpt-5-mini | OpenAI | $0.25 | $0.025 | 90% |
| gemini-3.1-flash-lite | Google Gemini | $0.25 | $0.025 | 90% |
| mistral-tiny | Mistral | $0.25 | $0.025 | 90% |
| deepseek-v3.2 | DeepSeek | $0.28 | $0.028 | 90% |
| gemini-2.5-flash | Google Gemini | $0.3 | $0.03 | 90% |
| gemini-3.5-flash-lite | Google Gemini | $0.3 | $0.03 | 90% |
| codestral-2508 | Mistral | $0.3 | $0.03 | 90% |
| open-mistral-nemo | Mistral | $0.3 | $0.03 | 90% |
| gemini-3-flash-preview | Google Gemini | $0.5 | $0.05 | 90% |
| mistral-large-3 | Mistral | $0.5 | $0.05 | 90% |
| mistral-large-2512 | Mistral | $0.5 | $0.05 | 90% |
| gpt-5.4-mini | OpenAI | $0.75 | $0.075 | 90% |
| gemini-3.6-flash | Google Gemini | $0.75 | $0.075 | 90% |
| gemini-3.7-flash | Google Gemini | $0.75 | $0.075 | 90% |
| gemini-3.8-flash | Google Gemini | $0.75 | $0.075 | 90% |
| claude-haiku-4-5 | Anthropic | $1 | $0.1 | 90% |
| gpt-5 | OpenAI | $1.25 | $0.125 | 90% |
| gpt-5.1 | OpenAI | $1.25 | $0.125 | 90% |
| gemini-2.5-pro | Google Gemini | $1.25 | $0.125 | 90% |
| gemini-3.5-flash | Google Gemini | $1.5 | $0.15 | 90% |
| mistral-medium | Mistral | $1.5 | $0.15 | 90% |
| mistral-medium-3 | Mistral | $1.5 | $0.15 | 90% |
| mistral-medium-3-5 | Mistral | $1.5 | $0.15 | 90% |
| mistral-medium-3.5 | Mistral | $1.5 | $0.15 | 90% |
| gpt-5.2 | OpenAI | $1.75 | $0.175 | 90% |
| gpt-5.6-terra | OpenAI | $2 | $0.2 | 90% |
| gpt-6-sol | OpenAI | $2 | $0.2 | 90% |
| claude-sonnet-5 | Anthropic | $2 | $0.2 | 90% |
| gemini-3.1-pro-preview | Google Gemini | $2 | $0.2 | 90% |
| gpt-5.4 | OpenAI | $2.5 | $0.25 | 90% |
| claude-sonnet-4-5 | Anthropic | $3 | $0.3 | 90% |
| claude-sonnet-4-6 | Anthropic | $3 | $0.3 | 90% |
| gpt-5.6 | OpenAI | $4 | $0.4 | 90% |
| gpt-5.6-sol | OpenAI | $4 | $0.4 | 90% |
| gpt-5.5 | OpenAI | $5 | $0.5 | 90% |
| claude-opus-4-5 | Anthropic | $5 | $0.5 | 90% |
| claude-opus-4-6 | Anthropic | $5 | $0.5 | 90% |
| claude-opus-4-7 | Anthropic | $5 | $0.5 | 90% |
| claude-opus-4-8 | Anthropic | $5 | $0.5 | 90% |
| claude-opus-5 | Anthropic | $5 | $0.5 | 90% |
| gpt-6-astra | OpenAI | $10 | $1 | 90% |
| claude-fable-5 | Anthropic | $10 | $1 | 90% |
| claude-mythos-5 | Anthropic | $10 | $1 | 90% |
| mistral-large-4 | Mistral | $1.36 | $0.14 | 90% |
| mistral-large-4-0 | Mistral | $1.36 | $0.14 | 90% |
| grok-4.5 | xAI | $2 | $0.3 | 85% |
| grok-4.3 | xAI | $1.25 | $0.2 | 84% |
| grok-4.20 | xAI | $1.25 | $0.2 | 84% |
| grok-4.20-0309-non-reasoning | xAI | $1.25 | $0.2 | 84% |
| grok-4.20-0309-reasoning | xAI | $1.25 | $0.2 | 84% |
| grok-4.20-non-reasoning | xAI | $1.25 | $0.2 | 84% |
| grok-4.20-reasoning | xAI | $1.25 | $0.2 | 84% |
| grok-build-0.1 | xAI | $1 | $0.2 | 80% |
| grok-code-fast | xAI | $1 | $0.2 | 80% |
| grok-code-fast-1-0825 | xAI | $1 | $0.2 | 80% |
| gpt-4.1-mini | OpenAI | $0.4 | $0.1 | 75% |
| gpt-4.1 | OpenAI | $2 | $0.5 | 75% |
| o3 | OpenAI | $2 | $0.5 | 75% |
| grok-4.6 | xAI | $2 | $0.5 | 75% |
| grok-4.7 | xAI | $2 | $0.5 | 75% |
| deepseek-r1 | DeepSeek | $0.55 | $0.14 | 75% |
| deepseek-v3 | DeepSeek | $0.27 | $0.07 | 74% |
| gpt-4o-mini | OpenAI | $0.15 | $0.075 | 50% |
| gpt-4o | OpenAI | $2.5 | $1.25 | 50% |
Some providers also charge for writing to the cache or for how long it is kept; check the provider's pricing page for those. To cache, put the part that repeats (instructions, examples, documents) at the start of the prompt.
Calculate with cached input →