AI API prices per 1M tokens
Input, output, cache and long-context prices per 1M tokens for 26 models from 11 providers’ own price lists, and the OpenRouter list for the other models we follow. USD; prices as listed in the United States on .
OpenAI API pricing
| Model | Input | Output | Cache read | Cache write | Long context, in / out | Checks |
|---|---|---|---|---|---|---|
| GPT-6 AstraOpenAI6 of 6 common tasks fit on price and context · quality not tested | $10.00 | $50.00 | $1.0010 % of input | $12.50×1.25 input | Above 272K: $20.00 / $75.00 | 25 checks on 25 dates |
| GPT-6 SolOpenAI6 of 6 common tasks fit on price and context · quality not tested | $2.00 | $10.00 | $0.2010 % of input | $2.50×1.25 input | Above 272K: $4.00 / $15.00 | 25 checks on 25 dates |
| GPT-6.1 SolOpenAI | $2.00 | $10.00 | $0.105 % of input | $2.50×1.25 input | Above 272K: $4.00 / $15.00 | 19 checks on 19 dates |
| GPT-5.4 MiniOpenAI6 of 6 common tasks fit on price and context · quality not tested | $0.75 | $4.50 | UnknownNot listed | UnknownNot listed | UnknownNot listed | 17 checks on 17 dates |
| GPT-6 LunaOpenAI6 of 6 common tasks fit on price and context · quality not tested | $0.10 | $0.50 | $0.0110 % of input | $0.125×1.25 input | Above 272K: $0.20 / $0.75 | 25 checks on 25 dates |
Published OpenAI’s price list, prices as listed in the United States on · USD per 1M tokens · source page
Claude API pricing (Anthropic)
| Model | Input | Output | Cache read | Cache write | Long context, in / out | Checks |
|---|---|---|---|---|---|---|
| Claude Fable 5.1Anthropic | $10.00 | $50.00 | $0.252.5 % of input | $12.50 5 min$20.00 1 hour | UnknownNot listed | 25 checks on 25 dates |
| Claude Opus 5.5Anthropic6 of 6 common tasks fit on price and context · quality not tested | $4.00 | $20.00 | $0.205 % of input | $5.00 5 min$8.00 1 hour | UnknownNot listed | 25 checks on 25 dates |
| Claude Sonnet 5.5Anthropic6 of 6 common tasks fit on price and context · quality not tested | $2.00 | $10.00 | $0.105 % of input | $2.50 5 min$4.00 1 hour | UnknownNot listed | 25 checks on 25 dates |
| Claude Haiku 5.5Anthropic6 of 6 common tasks fit on price and context · quality not tested | $0.10 | $0.50 | $0.0110 % of input | $0.125 5 min$0.20 1 hour | Above 100K: $0.50 / $2.50 | 22 checks on 22 dates |
Published Anthropic’s price list, prices as listed in the United States on · USD per 1M tokens · source page
Gemini API pricing (Google)
| Model | Input | Output | Cache read | Cache write | Long context, in / out | Checks |
|---|---|---|---|---|---|---|
| Gemini 3.1 Pro PreviewGoogle | $2.00 | $12.00 | $0.2010 % of input | UnknownNot listed | Above 200K: $4.00 / $18.00 | 17 checks on 17 dates |
| Gemini 4 ArgonGoogle6 of 6 common tasks fit on price and context · quality not tested | $2.00 | $10.00 | UnknownNot listed | UnknownNot listed | UnknownNot listed | 3 checks on 3 dates |
| Gemini 3.8 FlashGoogle6 of 6 common tasks fit on price and context · quality not tested | $0.75 | $3.75 | $0.07510 % of input | UnknownNot listed | UnknownNot listed | 17 checks on 17 dates |
Published Google’s price list, prices as listed in the United States on · USD per 1M tokens · source page
DeepSeek API pricing
| Model | Input | Output | Cache read | Cache write | Long context, in / out | Checks |
|---|---|---|---|---|---|---|
| DeepSeek V4 Pro 0813DeepSeek5 of 6 common tasks fit on price and context · quality not tested | $1.32 | $3.96 | $0.0443.3 % of input | UnknownNot listed | UnknownNot listed | 17 checks on 17 dates |
| DeepSeek V4.1 Flash 1017DeepSeek5 of 6 common tasks fit on price and context · quality not tested | $0.30 | $1.20 | $0.0062 % of input | UnknownNot listed | UnknownNot listed | 15 checks on 15 dates |
Published DeepSeek’s price list, prices as listed in the United States on · USD per 1M tokens · source page
Grok API pricing (xAI)
| Model | Input | Output | Cache read | Cache write | Long context, in / out | Checks |
|---|---|---|---|---|---|---|
| Grok 4.7xAI6 of 6 common tasks fit on price and context · quality not tested | $2.00 | $6.00 | $0.5025 % of input | UnknownNot listed | Above 200K: $4.00 / $12.00 | 25 checks on 25 dates |
Published xAI’s price list, prices as listed in the United States on · USD per 1M tokens · source page
Mistral API pricing
| Model | Input | Output | Cache read | Cache write | Long context, in / out | Checks |
|---|---|---|---|---|---|---|
| Mistral Medium 3.5Mistral | $1.50 | $7.50 | $0.1510 % of input | UnknownNot listed | UnknownNot listed | 10 checks on 10 dates |
| Mistral Large 4Mistral6 of 6 common tasks fit on price and context · quality not tested | $0.0710 % of input | UnknownNot listed | UnknownNot listed | 10 checks on 10 dates | ||
| Mistral Large 3 2512Mistral | $0.50 | $1.50 | $0.0510 % of input | UnknownNot listed | UnknownNot listed | 10 checks on 10 dates |
Published Mistral’s price list, prices as listed in the United States on · USD per 1M tokens · source page
Qwen API pricing (Alibaba Cloud)
| Model | Input | Output | Cache read | Cache write | Long context, in / out | Checks |
|---|---|---|---|---|---|---|
| Qwen3.8 Max (0902)Qwen6 of 6 common tasks fit on price and context · quality not tested | $2.00 | $6.00 | UnknownNot listed | UnknownNot listed | UnknownNot listed | 25 checks on 25 dates |
| Qwen3.8 FlashQwen5 of 6 common tasks fit on price and context · quality not tested | $0.15 | $0.47 | UnknownNot listed | UnknownNot listed | UnknownNot listed | 25 checks on 25 dates |
Published Alibaba Cloud’s price list, prices as listed in the United States on · USD per 1M tokens · source page
Open models on Groq
| Model | Input | Output | Cache read | Cache write | Long context, in / out | Checks |
|---|---|---|---|---|---|---|
| Qwen3.8 27BQwen | $0.80 | $4.00 | UnknownNot listed | UnknownNot listed | UnknownNot listed | 17 checks on 17 dates |
| gpt-oss-120bOpenAI | $0.15 | $0.60 | UnknownNot listed | UnknownNot listed | UnknownNot listed | 25 checks on 25 dates |
| gpt-oss-20bOpenAI | $0.075 | $0.30 | UnknownNot listed | UnknownNot listed | UnknownNot listed | 25 checks on 25 dates |
Published Groq’s price list, prices as listed in the United States on · USD per 1M tokens · source page
docs.z.ai price list
| Model | Input | Output | Cache read | Cache write | Long context, in / out | Checks |
|---|---|---|---|---|---|---|
| GLM 5.3Z.ai5 of 6 common tasks fit on price and context · quality not tested | $1.40 | $4.40 | $0.2618.6 % of input | UnknownNot listed | UnknownNot listed | 17 checks on 17 dates |
Published docs.z.ai’s price list, prices as listed in the United States on · USD per 1M tokens · source page
platform.kimi.ai price list
| Model | Input | Output | Cache read | Cache write | Long context, in / out | Checks |
|---|---|---|---|---|---|---|
| Kimi K3MoonshotAI4 of 6 common tasks fit on price and context · quality not tested | $3.00 | $15.00 | $0.3010 % of input | UnknownNot listed | UnknownNot listed | 15 checks on 15 dates |
Published platform.kimi.ai’s price list, prices as listed in the United States on · USD per 1M tokens · source page
platform.minimax.io price list
| Model | Input | Output | Cache read | Cache write | Long context, in / out | Checks |
|---|---|---|---|---|---|---|
| MiniMax M3MiniMax4 of 6 common tasks fit on price and context · quality not tested | $0.30 | $1.20 | UnknownNot listed | UnknownNot listed | UnknownNot listed | 17 checks on 17 dates |
Published platform.minimax.io’s price list, prices as listed in the United States on · USD per 1M tokens · source page
Other models we follow, on OpenRouter
| Model | Input | Output | Checks |
|---|---|---|---|
| Grok 4.5xAI | $2.00 | $6.00 | 25 checks on 25 dates |
Published OpenRouter models list, prices as listed in the United States on · USD per 1M tokens · every model on the list
“Not listed” — the price list does not publish that price, so we show none. Was → now marks a confirmed change in the last 7 days. Cache write: the price to store a prompt prefix; cache read: the price to reuse it. Common tasks: how many of the six tasks of the model picker fit on our data (published price, context window and answer length, and the task’s rules); quality on your task is not measured by Measured Lab yet.
Cheapest per task · Long-context cliff · Caching and batch · Rate limits by tier · What breaks when you switch · Rate limit planner · Task cost calculator · All models on OpenRouter · How we check prices