Models with Prompt caching

Models supporting cached prompt prefixes for cost reduction.

82 models

VisionTool useComputer useStreamingPrompt cachingBatch API

OpenAI

gpt-5.1-2025-11-13

99.7
Prompt caching ✓

OpenAI

gpt-4.1-mini-2025-04-14

99.7
Prompt caching ✓

OpenAI

gpt-4.1-2025-04-14

99.7
Prompt caching ✓

OpenAI

gpt-5.2-chat-latest

99.4
Prompt caching ✓Tier C

OpenAI

gpt-5.4-nano-2026-03-17

99.3
Prompt caching ✓

OpenAI

o3-mini-2025-01-31

99.3
Prompt caching ✓

OpenAI

gpt-5.5-2026-04-23

99.3
Prompt caching ✓

OpenAI

gpt-5.2

99.3
Prompt caching ✓Tier B

OpenAI

o1

99.3
Prompt caching ✓200K ctxTier C

OpenAI

o3-2025-04-16

99.3
Prompt caching ✓

OpenAI

o4-mini-2025-04-16

99.3
Prompt caching ✓

OpenAI

gpt-4o-2024-08-06

99.3
Prompt caching ✓Tier C

OpenAI

gpt-5.4-2026-03-05

99.3
Prompt caching ✓

OpenAI

gpt-3.5-turbo-0125

99.3
Prompt caching ✓

OpenAI

gpt-4-turbo-2024-04-09

99.3
Prompt caching ✓Tier C

OpenAI

gpt-4o

99.3
Prompt caching ✓128K ctxTier C

OpenAI

gpt-4o-mini

99.3
Prompt caching ✓128K ctxTier C

OpenAI

gpt-5-chat-latest

99.3
Prompt caching ✓Tier C

OpenAI

gpt-4-turbo

99.3
Prompt caching ✓128K ctxTier C

OpenAI

gpt-5.2-2025-12-11

99.3
Prompt caching ✓

OpenAI

gpt-4o-2024-11-20

99.3
Prompt caching ✓Tier C

OpenAI

gpt-4o-mini-2024-07-18

99.2
Prompt caching ✓Tier C

Anthropic

Claude Opus 4.6

99.1
Prompt caching ✓200K ctxTier B

OpenAI

gpt-5-search-api-2025-10-14

99.1
Prompt caching ✓

Google Gemini

Gemini 3 Flash Preview

99.1
Prompt caching ✓1.048576M ctxTier C

OpenAI

gpt-3.5-turbo

99.1
Prompt caching ✓Tier C

OpenAI

gpt-5.1

99.1
Prompt caching ✓Tier B

OpenAI

gpt-4o-search-preview

98.9
Prompt caching ✓Tier C

OpenAI

gpt-4o-mini-search-preview-2025-03-11

98.8
Prompt caching ✓

OpenAI

gpt-4o-mini-search-preview

98.8
Prompt caching ✓Tier C

OpenAI

gpt-5.3-chat-latest

98.7
Prompt caching ✓Tier C

OpenAI

gpt-5.4-mini

98.7
Prompt caching ✓Tier A

OpenAI

gpt-5.4-mini-2026-03-17

98.6
Prompt caching ✓

OpenAI

gpt-5.5

98.3
Prompt caching ✓Tier C

OpenAI

gpt-4o-2024-05-13

98.3
Prompt caching ✓Tier C

OpenAI

gpt-4.1-mini

98.1
Prompt caching ✓1.047576M ctxTier C

Anthropic

Claude Opus 4.7

98.0
Prompt caching ✓1M ctxTier B

OpenAI

gpt-5.1-chat-latest

98.0
Prompt caching ✓Tier C

OpenAI

gpt-5.4

98.0
Prompt caching ✓Tier A

OpenAI

gpt-4.1-nano-2025-04-14

97.8
Prompt caching ✓

OpenAI

o3

97.7
Prompt caching ✓200K ctxTier C

Anthropic

Claude Haiku 4.5

97.7
Prompt caching ✓200K ctxTier A

OpenAI

gpt-5-search-api

97.7
Prompt caching ✓Tier C

OpenAI

gpt-4

97.4
Prompt caching ✓Tier C

Anthropic

Claude Sonnet 4.6

97.4
Prompt caching ✓1M ctxTier A

Anthropic

Claude Opus 4.5

97.4
Prompt caching ✓200K ctxTier B

Google Gemini

Gemini Flash-Lite Latest

97.2
Prompt caching ✓1.048576M ctxTier C

Anthropic

Claude Sonnet 4.5

97.0
Prompt caching ✓200K ctxTier B

Google Gemini

Gemini 2.5 Flash-Lite

96.8
Prompt caching ✓1.048576M ctxTier B

OpenAI

gpt-3.5-turbo-1106

96.4
Prompt caching ✓

OpenAI

gpt-3.5-turbo-16k

96.0
Prompt caching ✓

OpenAI

gpt-4-0613

96.0
Prompt caching ✓

Anthropic

Claude Opus 4.1

95.1
Prompt caching ✓200K ctxTier C

OpenAI

gpt-4o-search-preview-2025-03-11

93.3
Prompt caching ✓

Google Gemini

Gemini 3.1 Flash Lite

93.3
Prompt caching ✓1.048576M ctx

OpenAI

o4-mini

93.0
Prompt caching ✓Tier C

OpenAI

gpt-4.1

92.8
Prompt caching ✓1.047576M ctxTier B

OpenAI

gpt-4.1-nano

90.0
Prompt caching ✓1.047576M ctxTier C

Anthropic

Claude Opus 4.8

89.1
Prompt caching ✓1M ctxTier A

OpenAI

gpt-5.4-nano

88.1
Prompt caching ✓Tier C

OpenAI

o1-2024-12-17

81.4
Prompt caching ✓Tier C

OpenAI

gpt-5-mini-2025-08-07

81.3
Prompt caching ✓

OpenAI

gpt-5-mini

80.7
Prompt caching ✓Tier C

Google Gemini

Gemini Pro Latest

80.2
Prompt caching ✓1.048576M ctxTier C

Google Gemini

Gemini Flash Latest

80.0
Prompt caching ✓1.048576M ctxTier B

Google Gemini

Gemini 3.5 Flash

78.0
Prompt caching ✓1.048576M ctxTier A

OpenAI

o3-mini

66.2
Prompt caching ✓200K ctxTier C

Google Gemini

Gemini 3.1 Pro Preview

62.9
Prompt caching ✓1.048576M ctxTier C

OpenAI

gpt-5

48.3
Prompt caching ✓Tier C

Google Gemini

Gemini 2.5 Flash

44.9
Prompt caching ✓1.048576M ctxTier A

OpenAI

gpt-5-2025-08-07

37.2
Prompt caching ✓

OpenAI

gpt-5-nano-2025-08-07

31.7
Prompt caching ✓

OpenAI

gpt-5-nano

28.9
Prompt caching ✓Tier C

Google Gemini

Gemini 2.5 Pro

27.2
Prompt caching ✓1.048576M ctxTier A

Google Gemini

Deep Research Pro Preview (Dec-12-2025)

no data
Prompt caching ✓131K ctx

OpenRouter

DeepSeek v3.2

no data
Prompt caching ✓131K ctxTier A

Google Gemini

Gemini 2.5 Pro Preview TTS

no data
Prompt caching ✓8K ctx

Google Gemini

Gemini 3.1 Pro Preview Custom Tools

no data
Prompt caching ✓1.048576M ctxTier C

OpenRouter

MiniMax M2.5

no data
Prompt caching ✓256K ctxTier A

Google Gemini

Nano Banana

no data
Prompt caching ✓33K ctx

Google Gemini

Nano Banana 2

no data
Prompt caching ✓66K ctx

Google Gemini

Nano Banana Pro

no data
Prompt caching ✓131K ctx