Skip to main content
Watch: I Built an AI App Without Leaving Databricks

API capability - Prompt caching

AI models with prompt caching

136 of the 167 models in the registry have prompt caching, according to each provider's own docs. The cheapest by output price is Ministral 3 3B from Mistral AI at $0.10 per 1M output tokens. The largest context window among them is 1.05M tokens, on GPT-6 Luna.

Capabilities checked . Sources

Models with Yes
136 of 167
Lowest output / 1M
$0.10
Largest context
1.05M
Not stated
18

Every model with prompt caching, cheapest first

Models with prompt caching, sorted by output price, with input and output price per 1M tokens, context window and the provider doc page that states it.
ModelProviderInput / 1MOutput / 1MContextSource
Ministral 3 3BMistral AI$0.10$0.10256Kdocs.mistral.ai
Qwen3.7-FlashAlibaba Cloud (Qwen)$0.03$0.131Malibabacloud.com
Amazon Nova MicroAmazon$0.035$0.14128Kdocs.aws.amazon.com
Ministral 3 8BMistral AI$0.15$0.15256Kdocs.mistral.ai
Ministral 3 14BMistral AI$0.20$0.20256Kdocs.mistral.ai
Amazon Nova LiteAmazon$0.06$0.24300Kdocs.aws.amazon.com
MiMo-V2.6-FlashXiaomi MiMo$0.14$0.281Mmimo.mi.com
Step 3.5 FlashStepFun$0.10$0.30256Kplatform.stepfun.ai
Step 3.5 Flash 2603StepFun$0.10$0.30256Kplatform.stepfun.ai
GPT-5 nanoOpenAI$0.05$0.40400Kdevelopers.openai.com
Qwen-FlashAlibaba Cloud (Qwen)$0.05$0.401Malibabacloud.com
GLM-4.7-FlashXZ.ai$0.07$0.40200Kdocs.z.ai
Gemini 2.5 Flash-LiteGoogle$0.10$0.401.05Mai.google.dev
GPT-4.1 nanoOpenAI$0.10$0.401.05Mdevelopers.openai.com
Qwen3.5-FlashAlibaba Cloud (Qwen)$0.10$0.401Malibabacloud.com
Seed 2.0 MiniByteDance Seed$0.10$0.40256Kdocs.byteplus.com
ERNIE 4.5 Turbo (128K)Baidu$0.11$0.45128Kintl.cloud.baidu.com
ERNIE 4.5 Turbo (32K)Baidu$0.11$0.4532Kintl.cloud.baidu.com
Qwen3.8-FlashAlibaba Cloud (Qwen)$0.15$0.471Malibabacloud.com
Qwen3.8-Omni-FlashAlibaba Cloud (Qwen)$0.15$0.471Malibabacloud.com
Claude Haiku 5.5Anthropic$0.10$0.501Mplatform.claude.com
GPT-6 LunaOpenAI$0.10$0.501.05Mdevelopers.openai.com
GLM-5.3-FlashZ.ai$0.15$0.501Mdocs.z.ai
Hy3Tencent$0.132$0.528256Ktencentcloud.com
GPT-4o MiniOpenAI$0.15$0.60128Kdevelopers.openai.com
Mistral Small 4Mistral AI$0.15$0.60256Kdocs.mistral.ai
Mercury 2.5Inception$0.20$0.75260Kdocs.inceptionlabs.ai
Mercury 2Inception$0.25$0.75128Kdocs.inceptionlabs.ai
Mercury Edit 2Inception$0.25$0.7532Kdocs.inceptionlabs.ai
MiMo-V2.6-ProXiaomi MiMo$0.435$0.871Mmimo.mi.com
Codestral 25.08Mistral AI$0.30$0.90128Kdocs.mistral.ai
GLM-4.6VZ.ai$0.30$0.90128Kdocs.z.ai
GLM-4.5-AirZ.ai$0.20$1.10128Kdocs.z.ai
Step 3.7 FlashStepFun$0.20$1.15256Kplatform.stepfun.ai
Step-1o Turbo VisionStepFun$0.36$1.1532Kplatform.stepfun.ai
GPT-5.6 LunaOpenAI$0.20$1.201.05Mdevelopers.openai.com
DeepSeek V4.1 FlashDeepSeek$0.30$1.201Mapi-docs.deepseek.com
MiniMax-M2.5MiniMax$0.30$1.20205Kplatform.minimax.io
MiniMax-M2.7MiniMax$0.30$1.20205Kplatform.minimax.io
MiniMax-M3MiniMax$0.30$1.201Mplatform.minimax.io
Qwen-PlusAlibaba Cloud (Qwen)$0.40$1.201Malibabacloud.com
GPT-5.4 nanoOpenAI$0.20$1.25400Kdevelopers.openai.com
GLM-5.3-FlashXZ.ai$0.37$1.251Mdocs.z.ai
ERNIE 4.5 Turbo VLBaidu$0.42$1.27128Kintl.cloud.baidu.com
ERNIE 4.5 Turbo VL (32K)Baidu$0.42$1.2732Kintl.cloud.baidu.com
Gemini 3.1 Flash-LiteGoogle$0.25$1.501.05Mai.google.dev
Qwen3.6-FlashAlibaba Cloud (Qwen)$0.25$1.501Malibabacloud.com
Qwen3-Coder-FlashAlibaba Cloud (Qwen)$0.30$1.501Malibabacloud.com
Mistral Large 3Mistral AI$0.50$1.50256Kdocs.mistral.ai
Qwen3-VL-PlusAlibaba Cloud (Qwen)$0.20$1.60262Kalibabacloud.com
GPT-4.1 MiniOpenAI$0.40$1.601.05Mdevelopers.openai.com
Qwen3.7-PlusAlibaba Cloud (Qwen)$0.40$1.601Malibabacloud.com
GPT-5 MiniOpenAI$0.25$2400Kdevelopers.openai.com
Seed 2.0 LiteByteDance Seed$0.25$2256Kdocs.byteplus.com
Grok Build 0.1xAI$1$2256Kdocs.x.ai
GLM-4.5Z.ai$0.60$2.20128Kdocs.z.ai
GLM-4.6Z.ai$0.60$2.20200Kdocs.z.ai
GLM-4.7Z.ai$0.60$2.20200Kdocs.z.ai
Qwen3.5-PlusAlibaba Cloud (Qwen)$0.40$2.401Malibabacloud.com
MiniMax-M2.5-highspeedMiniMax$0.60$2.40205Kplatform.minimax.io
MiniMax-M2.7-highspeedMiniMax$0.60$2.40205Kplatform.minimax.io
Amazon Nova 2 LiteAmazon$0.30$2.501Mdocs.aws.amazon.com
Gemini 2.5 FlashGoogle$0.30$2.501.05Mai.google.dev
Gemini 3.5 Flash-LiteGoogle$0.30$2.501.05Mai.google.dev
Dola Seed 2.1 TurboByteDance Seed$0.50$2.50256Kdocs.byteplus.com
Grok 4.20xAI$1.25$2.501Mdocs.x.ai
Grok 4.20 (Non-Reasoning)xAI$1.25$2.501Mdocs.x.ai
Grok 4.20 Multi-Agent BetaxAI$1.25$2.501Mdocs.x.ai
Grok 4.3xAI$1.25$2.501Mdocs.x.ai
Hy4 previewTencent$0.834$2.5011Mtencentcloud.com
Step 5 PreviewStepFun$1$2.701Mplatform.stepfun.ai
Gemini 3 Flash PreviewGoogle$0.50$31.05Mai.google.dev
Qwen3.6-PlusAlibaba Cloud (Qwen)$0.50$31Malibabacloud.com
Qwen3.8-27BAlibaba Cloud (Qwen)$0.50$31Malibabacloud.com
Seed 2.0 ProByteDance Seed$0.50$3256Kdocs.byteplus.com
Amazon Nova ProAmazon$0.80$3.20300Kdocs.aws.amazon.com
GLM-5Z.ai$1$3.20200Kdocs.z.ai
Gemini 3.6 FlashGoogle$0.75$3.751.05Mai.google.dev
Gemini 3.8 FlashGoogle$0.75$3.751.05Mai.google.dev
DeepSeek V4 ProDeepSeek$1.32$3.961Mapi-docs.deepseek.com
Kimi K2.6Moonshot AI$0.95$4262Kplatform.kimi.ai
Kimi K2.7 CodeMoonshot AI$0.95$4262Kplatform.kimi.ai
Mistral Large 4Mistral AI$1.36$4.181Mdocs.mistral.ai
o3-miniOpenAI$1.10$4.40200Kdevelopers.openai.com
o4-miniOpenAI$1.10$4.40200Kdevelopers.openai.com
GLM-5.1Z.ai$1.40$4.40200Kdocs.z.ai
GLM-5.2Z.ai$1.40$4.401Mdocs.z.ai
GLM-5.3Z.ai$1.40$4.401Mdocs.z.ai
GPT-5.4 MiniOpenAI$0.75$4.50400Kdevelopers.openai.com
Claude Haiku 4.5Anthropic$1$5200Kplatform.claude.com
Gemini Robotics ER 2 PreviewGoogle$1$5131Kai.google.dev
Grok 4.5xAI$2$6500Kdocs.x.ai
Grok 4.6xAI$2$6500Kdocs.x.ai
Grok 4.7xAI$2$6500Kdocs.x.ai
Qwen3.8-2.4T-A95BAlibaba Cloud (Qwen)$2$61Malibabacloud.com
Qwen3.8-MaxAlibaba Cloud (Qwen)$2$61Malibabacloud.com
Qwen-MaxAlibaba Cloud (Qwen)$1.60$6.4033Kalibabacloud.com
Mistral Medium 3.5Mistral AI$1.50$7.50256Kdocs.mistral.ai
Qwen3.7-MaxAlibaba Cloud (Qwen)$2.50$7.501Malibabacloud.com
Kimi K2.7 Code HighSpeedMoonshot AI$1.90$8262Kplatform.kimi.ai
GPT-4.1OpenAI$2$81.05Mdevelopers.openai.com
o3OpenAI$2$8200Kdevelopers.openai.com
Gemini 2.5 ProGoogle$1.25$101.05Mai.google.dev
GPT-5OpenAI$1.25$10400Kdevelopers.openai.com
GPT-5.1OpenAI$1.25$10400Kdevelopers.openai.com
Claude Sonnet 5Anthropic$2$101Mplatform.claude.com
Claude Sonnet 5.5Anthropic$2$101Mplatform.claude.com
Gemini 4 ArgonGoogle$2$101.05Mai.google.dev
GPT-6 SolOpenAI$2$101.05Mdevelopers.openai.com
GPT-6.1 SolOpenAI$2$101.05Mdevelopers.openai.com
GPT-4oOpenAI$2.50$10128Kdevelopers.openai.com
Gemini 3.1 Pro PreviewGoogle$2$121.05Mai.google.dev
GPT-5.6 TerraOpenAI$2$121.05Mdevelopers.openai.com
GPT-5.2OpenAI$1.75$14400Kdevelopers.openai.com
GPT-5.3-CodexOpenAI$1.75$14400Kdevelopers.openai.com
GPT-5.4OpenAI$2.50$151.05Mdevelopers.openai.com
Claude Sonnet 4.5Anthropic$3$15200Kplatform.claude.com
Claude Sonnet 4.6Anthropic$3$151Mplatform.claude.com
Kimi K3Moonshot AI$3$151.05Mplatform.kimi.ai
Claude Opus 5.5Anthropic$4$201Mplatform.claude.com
GPT-5.6 SolOpenAI$4$201.05Mdevelopers.openai.com
Claude Opus 4.5Anthropic$5$25200Kplatform.claude.com
Claude Opus 4.6Anthropic$5$251Mplatform.claude.com
Claude Opus 4.7Anthropic$5$251Mplatform.claude.com
Claude Opus 4.8Anthropic$5$251Mplatform.claude.com
Claude Opus 5Anthropic$5$251Mplatform.claude.com
Chat LatestOpenAI$5$30400Kdevelopers.openai.com
GPT-5.5OpenAI$5$301.05Mdevelopers.openai.com
Claude Fable 5Anthropic$10$501Mplatform.claude.com
Claude Fable 5.1Anthropic$10$501Mplatform.claude.com
Claude Mythos 5Anthropic$10$501Mplatform.claude.com
Claude Mythos 5.1Anthropic$10$501Mplatform.claude.com
GPT-6 AstraOpenAI$10$501.05Mdevelopers.openai.com
o1OpenAI$15$60200Kdevelopers.openai.com
GPT-5.6 CyberOpenAI$12.50$75400Kdevelopers.openai.com
GPT-5.5 ProOpenAI$30$1801.05Mdevelopers.openai.com

Prices in USD per 1M tokens at the standard rate, sorted by output price. 13 models are documented without it, and for 18 models the provider's docs do not say (Not stated).

FAQ

What is the cheapest model with prompt caching?

Ministral 3 3B from Mistral AI is the cheapest model with prompt caching in the registry as of October 18, 2026, at $0.10 per 1M input tokens and $0.10 per 1M output tokens on the provider's own API.

Which model with prompt caching has the largest context window?

GPT-6 Luna from OpenAI has the largest context window among models with prompt caching: 1.05M tokens, at $0.50 per 1M output tokens.

How many AI models have prompt caching?

136 of the 167 models in the registry have prompt caching. 13 models do not, and for 18 models the provider's docs do not say either way.

What counts as prompt caching?

The provider offers prompt caching for the model, so a repeated prompt prefix can be reused instead of processed again from scratch. Each Yes on this page links to the provider's own documentation page that states it.

Sources

Every Yes links to the provider doc page that states it, checked on . Prices and context windows come from each model's pricing page; open a model for its rate card and pricing sources.

Documentation pages read (75)

Newsletter

Get model price changes as they ship

One email per week. New model releases, price cuts and what they mean for your bill.