Prompt caching pricing & savings

When many requests start with the same long prefix (system prompt, tool definitions, a shared document), providers can cache it and bill those tokens at a much lower cached-input rate. Chatbots and agent loops benefit most because they resend the same context every turn.

Estimate your monthly bill with this discount →

Model	Normal input /1M	Cached input /1M	Saving
DeepSeek V4 Flash	$0.30	$0.006	98%
Claude Fable 5.1	$10.00	$0.25	98%
Claude Mythos 5.1	$10.00	$0.25	98%
DeepSeek V4 Pro	$1.32	$0.044	97%
Claude Opus 5.5	$4.00	$0.20	95%
GPT-6.1 Sol	$2.00	$0.10	95%
Claude Sonnet 4.5	$3.00	$0.30	90%
Claude Sonnet 4.6	$3.00	$0.30	90%
Gemini 2.5 Flash Lite	$0.10	$0.01	90%
Gemini 3.5 Flash	$1.50	$0.15	90%
Gemini 3.6 Flash	$0.75	$0.075	90%
Gemini 3.7 Flash	$0.75	$0.075	90%
Gemini 3.8 Flash	$0.75	$0.075	90%
Gemini Flash (latest)	$0.75	$0.075	90%
Devstral Medium (latest)	$0.40	$0.04	90%
Devstral Small (latest)	$0.10	$0.01	90%
Magistral Medium (latest)	$1.50	$0.15	90%
Ministral 14b (latest)	$0.20	$0.02	90%
Ministral 3b (latest)	$0.10	$0.01	90%
Mistral Medium 3	$1.50	$0.15	90%
Mistral Medium 3.5	$1.50	$0.15	90%
GPT-5 nano	$0.05	$0.005	90%
GPT-5.2	$1.75	$0.175	90%
GPT-5.4 mini	$0.75	$0.075	90%
GPT-5.4 nano	$0.20	$0.02	90%
GPT-5.6 Luna	$0.20	$0.02	90%
GPT-6 Luna	$0.10	$0.01	90%
Claude Fable 5	$10.00	$1.00	90%
Claude Haiku 4.5	$1.00	$0.10	90%
Claude Mythos 5	$10.00	$1.00	90%
Claude Opus 4.5	$5.00	$0.50	90%
Claude Opus 4.6	$5.00	$0.50	90%
Claude Opus 4.7	$5.00	$0.50	90%
Claude Opus 4.8	$5.00	$0.50	90%
Claude Opus 5	$5.00	$0.50	90%
Claude Sonnet 5	$2.00	$0.20	90%
Claude Sonnet 5.5	$2.00	$0.20	90%
Gemini 2.5 Flash	$0.30	$0.03	90%
Gemini 2.5 Pro	$1.25	$0.125	90%
Gemini 3 Flash Preview	$0.50	$0.05	90%
Gemini 3.1 Flash Lite	$0.25	$0.025	90%
Gemini 3.1 Pro Preview	$2.00	$0.20	90%
Gemini 3.5 Flash Lite	$0.30	$0.03	90%
Gemini Flash Lite (latest)	$0.30	$0.03	90%
Gemini Pro (latest)	$2.00	$0.20	90%
Codestral (latest)	$0.30	$0.03	90%
Magistral Small (latest)	$0.15	$0.015	90%
Ministral 8b (latest)	$0.15	$0.015	90%
Mistral Large (latest)	$0.50	$0.05	90%
Mistral Large 4	$0.68	$0.068	90%
Mistral Small (latest)	$0.15	$0.015	90%
GPT-5	$1.25	$0.125	90%
GPT-5 mini	$0.25	$0.025	90%
GPT-5.1	$1.25	$0.125	90%
GPT-5.4	$2.50	$0.25	90%
GPT-5.5	$5.00	$0.50	90%
GPT-5.6 Sol	$4.00	$0.40	90%
GPT-5.6 Terra	$2.00	$0.20	90%
GPT-6 Astra	$10.00	$1.00	90%
GPT-6 Sol	$2.00	$0.20	90%
Grok 4.5	$2.00	$0.30	85%
Grok 4.20 Non Reasoning	$1.25	$0.20	84%
Grok 4.20 Reasoning	$1.25	$0.20	84%
Grok 4.3	$1.25	$0.20	84%
Grok Build 0.1	$1.00	$0.20	80%
Grok Code Fast 1	$1.00	$0.20	80%
GPT-4.1	$2.00	$0.50	75%
GPT-4.1 mini	$0.40	$0.10	75%
o3	$2.00	$0.50	75%
Grok 4.6	$2.00	$0.50	75%
Grok 4.7	$2.00	$0.50	75%
GPT-4o	$2.50	$1.25	50%
GPT-4o mini	$0.15	$0.075	50%

FAQ

Does writing to the cache cost extra?

Anthropic charges a premium (shown as cache write on model pages) the first time a prefix is cached; OpenAI and Google cache automatically without a write fee in most cases.

How long does a cached prompt last?

Typically minutes (often around 5) unless refreshed by new requests; some providers sell longer cache lifetimes.

What cache hit rate is realistic?

Agent loops and multi-turn chats often reach 50–80% of input tokens; one-off requests with unique documents get close to 0%.