Cheapest cached-input pricing
Models with explicit cache-read pricing — the single biggest lever for high-volume apps that resend the same system prompt.
Current leader: Nemotron 3.5 Lightning 30b A3b (Perplexity). Refreshed 2026-10-28 12:57 UTC.
| # | Model | Provider | Input $/1M | Output $/1M | Cached $/1M | Context |
|---|---|---|---|---|---|---|
| 1 | Nemotron 3.5 Lightning 30b A3b | Perplexity | $0.011 | $0.17 | $0.0011 | — |
| 2 | Muse Spark 1.2 Contributor | Meta | $0.1 | $0.2 | $0.002 | 1.0M |
| 3 | Muse Spark 1.3 Contributor | Meta | $0.1 | $0.2 | $0.002 | 1.0M |
| 4 | Mimo V2.5 | Novita AI | $0.168 | $0.336 | $0.0034 | 1.0M |
| 5 | DeepSeek V4.1 Flash | DeepInfra | $0.14 | $0.42 | $0.0042 | 1.0M |
| 6 | DeepSeek V4.1 Flash 1017 | DeepInfra | $0.14 | $0.42 | $0.0042 | 1.0M |
| 7 | Mimo V2.5 Pro | Novita AI | $0.522 | $1.04 | $0.0043 | 1.0M |
| 8 | Databricks GPT 5 Nano | Databricks | $0.05 | $0.4 | $0.005 | 272K |
| 9 | GPT 5 Nano | OpenAI | $0.05 | $0.4 | $0.005 | 272K |
| 10 | GPT 5 Nano | OpenAI | $0.05 | $0.4 | $0.005 | 272K |
| 11 | GPT 5 Nano | Perplexity | $0.05 | $0.4 | $0.005 | — |
| 12 | GPT 5 Nano | Azure OpenAI | $0.05 | $0.4 | $0.005 | 272K |
| 13 | GPT 5 Nano | Azure OpenAI | $0.055 | $0.44 | $0.0055 | 272K |
| 14 | Deepseek Flash | DeepSeek | $0.3 | $1.2 | $0.006 | 1M |
| 15 | Deepseek V4 Flash | DeepSeek | $0.3 | $1.2 | $0.006 | 1M |
| 16 | Deepseek V4 Flash Vision Exp | DeepSeek | $0.3 | $1.2 | $0.006 | 1M |
| 17 | Deepseek V4p1 Flash | Fireworks AI | $0.3 | $1.2 | $0.006 | 1.0M |
| 18 | Deepseek V4p1 Flash 1017 | Fireworks AI | $0.3 | $1.2 | $0.006 | 1.0M |
| 19 | DeepSeek V4.1 Flash | Together AI | $0.3 | $1.2 | $0.006 | 1.0M |
| 20 | DeepSeek V4.1 Flash 1017 | Together AI | $0.3 | $1.2 | $0.006 | 1.0M |
| 21 | DeepSeek V4.1 Flash | Baseten | $0.3 | $1.2 | $0.007 | 1.0M |
| 22 | DeepSeek V4.1 Flash 1017 | Baseten | $0.3 | $1.2 | $0.007 | 1.0M |
| 23 | Deepseek V4 Flash 0731 | Fireworks AI | $0.22 | $0.66 | $0.007 | 1.0M |
| 24 | Deepseek V4 Flash Vision Exp | Fireworks AI | $0.22 | $0.66 | $0.007 | 1.0M |
| 25 | Deepseek V4p1 Flash Us | Fireworks AI | $0.45 | $1.8 | $0.009 | 1.0M |
| 26 | Deepseek V4p1 Flash 1017 Us | Fireworks AI | $0.45 | $1.8 | $0.009 | 1.0M |
| 27 | Claude Haiku 5 5 | Anthropic | $0.1 | $0.5 | $0.01 | 1M |
| 28 | Gemini 2.5 Flash Lite | Google Gemini | $0.1 | $0.4 | $0.01 | 1.0M |
| 29 | Devstral Small | Mistral AI | $0.1 | $0.3 | $0.01 | 256K |
| 30 | Ministral 3b 2512 | Mistral AI | $0.1 | $0.1 | $0.01 | 131K |
| 31 | Ministral 3b | Mistral AI | $0.1 | $0.1 | $0.01 | 131K |
| 32 | Voxtral Small 2507 | Mistral AI | $0.1 | $0.4 | $0.01 | 33K |
| 33 | Voxtral Small | Mistral AI | $0.1 | $0.4 | $0.01 | 33K |
| 34 | Mistral Small | Mistral AI | $0.1 | $0.3 | $0.01 | 32K |
| 35 | Ministral 3 3b 2512 | Mistral AI | $0.1 | $0.1 | $0.01 | 131K |
| 36 | GPT 6 Luna | OpenAI | $0.1 | $0.5 | $0.01 | 922K |
| 37 | GPT 6 Luna | Perplexity | $0.1 | $0.5 | $0.01 | — |
| 38 | Claude Haiku 5 5 | Perplexity | $0.1 | $0.5 | $0.01 | — |
| 39 | GPT 6 Luna | Azure OpenAI | $0.1 | $0.5 | $0.01 | 922K |
| 40 | GPT 6 Luna | Azure OpenAI | $0.1 | $0.5 | $0.01 | 922K |
| 41 | Claude Haiku 5 5 | Amazon Bedrock | $0.1 | $0.5 | $0.01 | 1M |
| 42 | GLM 4.7 Flash | DeepInfra | $0.06 | $0.4 | $0.01 | 203K |
| 43 | Nemotron Lightning 3p5 30b A3b | Fireworks AI | $0.05 | $0.2 | $0.01 | 262K |
| 44 | Glm 4.7 Flash | Novita AI | $0.07 | $0.4 | $0.01 | 200K |
| 45 | Amazon.nova Micro V1:0 | Amazon Bedrock | $0.042 | $0.168 | $0.011 | 128K |
| 46 | Anthropic.claude Haiku 5 5 | Amazon Bedrock | $0.12 | $0.6 | $0.012 | 1M |
| 47 | Ling 3.0 Flash | DeepInfra | $0.06 | $0.18 | $0.012 | 131K |
| 48 | Ling 3.0 Flash Fast | Novita AI | $0.06 | $0.18 | $0.012 | 262K |
| 49 | Ling 3.0 Flash | Novita AI | $0.06 | $0.18 | $0.012 | 262K |
| 50 | Deepseek Coder | DeepSeek | $0.14 | $0.28 | $0.014 | 128K |
| 51 | DeepSeek V4.1 Flash Fast | Baseten | $0.6 | $2.4 | $0.014 | 1.0M |
| 52 | DeepSeek V4.1 Flash 1017 Fast | Baseten | $0.6 | $2.4 | $0.014 | 1.0M |
| 53 | Magistral Small | Mistral AI | $0.15 | $0.6 | $0.015 | 262K |
| 54 | Mistral Small | Mistral AI | $0.15 | $0.6 | $0.015 | 262K |
| 55 | Ministral 3 8b 2512 | Mistral AI | $0.15 | $0.15 | $0.015 | 262K |
| 56 | Ministral 8b 2512 | Mistral AI | $0.15 | $0.15 | $0.015 | 262K |
| 57 | Ministral 8b | Mistral AI | $0.15 | $0.15 | $0.015 | 262K |
| 58 | Mistral Small 2603 | Mistral AI | $0.15 | $0.6 | $0.015 | 262K |
| 59 | Mistral Vibe Cli Fast | Mistral AI | $0.15 | $0.6 | $0.015 | 262K |
| 60 | GPT Oss 120b | Fireworks AI | $0.15 | $0.6 | $0.015 | 131K |
Other use cases
Frequently asked
› Which provider has the cheapest prompt caching?
Nemotron 3.5 Lightning 30b A3b from Perplexity currently leads this list, at $0.0115/1M input tokens. Rankings are recomputed from the raw dataset on every refresh, so they track price cuts automatically.
› How are these rankings calculated?
Each use case applies a capability filter (context size, vision, tool calling, cache pricing) and then ranks the surviving models by the blended cost that profile actually pays. See the methodology page for the exact formulas.
› Is the cheapest model always the right choice?
No. Check the capability flags and context window, and benchmark quality on your own data — the cheapest model that fails your evals costs more than the second cheapest that passes them.