GPT-6 Luna
工具调用推理缓存结构化输出
官方价格
$0.10 / $0.50
输入 / 输出 · OpenAI
最低输入价
$0.0137
KFCV50
上下文
1.1M
比较服务商
每个点是一家服务商。从左到右是每百万 token 的输入价,从下到上是输出价。虚线是官方价格。又左又下的点两边都更便宜。 离其余报价太远的价格没有画在这张图上。
基础价格
6最新价格检查于 Nov 18, 2026。
服务商价格
13金额均以美元计价。悬停或点击带下划线的金额,可查看价格换算明细。
DR 是服务商网站的 Ahrefs 域名评级,范围 0 到 100。域名年龄是该网站的注册时间。这些只说明公开痕迹,不能说明服务商是否会持续运营。域名评级低于 10 的报价不列出。
未验证报价(5)
金额均以美元计价。悬停或点击带下划线的金额,可查看价格换算明细。
DR 是服务商网站的 Ahrefs 域名评级,范围 0 到 100。域名年龄是该网站的注册时间。这些只说明公开痕迹,不能说明服务商是否会持续运营。域名评级低于 10 的报价不列出。
GPT-6 Luna · 已公布价格
| Standard tier | Input | Output | Cache read |
|---|---|---|---|
| Up to 272K | $0.10 | $0.50 | $0.01 |
| Over 272K | $0.20 | $0.75 | $0.02 |
GPT-6 Luna · 权重许可证
| 模型权重 | 协议 |
|---|---|
| 闭源 / 专有 | 闭源 / 专有 |
缓存与处理模式
GPT 6 Luna price · Short-context Standard · USD per 1M tokens
Cached input is $0.01 per million tokens, and cache writes are $0.125.
[2]For prompts above 272,000 input tokens, input and cache rates double and the output rate increases by 50 percent, applied to the full request.
[2]USD per 1M tokens · short-context Standard reference
Batch and Flex cost half of Standard.
[2]
关于 GPT-6 Luna
Yes. Above 272,000 input tokens, input and cache rates double and the output rate increases by 50 percent for the full request. Short-context Standard is $0.10 input and $0.50 output per million tokens. Long-context Standard is $0.20 input and $0.75 output. Batch and Flex are half of the matching Standard row, Fast is twice that row, and regional processing adds 10 percent where available.
Short-context Standard cached input is $0.01 per million tokens, and the cache write is $0.125. Above 272,000 input tokens those cache rates double, to $0.02 and $0.25. Output is unchanged by the cache: it stays on the output rate for that context tier.
On short-context Standard, Luna is $0.10 input and $0.50 output per million tokens, and Sol is $2 and $10. Cached input is $0.01 versus $0.20, and cache writes are $0.125 versus $2.50. Both use the same 272,000-token rule and the same Batch, Flex, Fast, and regional multipliers. Compare matching modes. A Luna Fast quote is not a discount against Sol Standard.
For GPT 6 Luna cost, use the input, output, and cached-input rates shown in the official rate table for your prompt length and processing mode. A consumer subscription does not replace this API bill.
来源与注脚
- [1]来源核验日期:
- [2]
- [3]
输入、输出和缓存读取的单位是每百万 token 的美元价格。速度是每秒输出 token 数。可用率是来源报告的在线比例。









