Site descriptions, notices, and reviews are shown in the language in which they were submitted.
HI API 是面向开发者的统一 AI 网关，通过一个兼容 OpenAI 协议的端点，接入全球主流大模型，统一计费、统一监控。 极简接入： 无需复杂改动，改个域名与 Key 即可无缝切换。 极致稳定： 智能路由与多节点冗余，保障 99.9% 的高可用性与低延迟响应。 透明计费： 按量付费，实时消耗明细，拒绝任何隐形消费。
Price listing
Fable 5
2 public records are listed; open the full page to review details as needed.
Price listing
Opus 4.8
2 public records are listed; open the full page to review details as needed.
Price listing
Opus 4.7
2 public records are listed; open the full page to review details as needed.
Price listing
Sonnet 5
2 public records are listed; open the full page to review details as needed.
Price listing
GPT 5.6 Sol
2 public records are listed; open the full page to review details as needed.
Price listing
GPT 5.6 Terra
1 public records are listed; open the full page to review details as needed.
Price listing
GLM 5.2
3 public records are listed; open the full page to review details as needed.
Price listing
GPT Image 2
1 public records are listed; open the full page to review details as needed.
Price listing
Opus 4.6
2 public records are listed; open the full page to review details as needed.
Current public prices are shown by model and channel. Token prices use CNY per 1M tokens unless another billing unit is shown.
| Channel | Provider model | Current price | Price trend | Uptime | Fake % | Latency |
|---|---|---|---|---|---|---|
| Claude Kiro高缓 | claude-fable-5 | 输入 ¥0.9 / 输出 ¥4.5 / 缓存 ¥0.09 / 写入 ¥1.13 | -10% | — | — | — |
| 满血Claude【禁蒸馏】 | claude-fable-5 | 输入 ¥8 / 输出 ¥40 / 缓存 ¥0.8 / 写入 ¥10 | 0% | — | — | — |
| Channel | Provider model | Current price | Price trend | Uptime | Fake % | Latency |
|---|---|---|---|---|---|---|
| Claude Kiro高缓 | claude-opus-4-8 | 输入 ¥0.45 / 输出 ¥2.25 / 缓存 ¥0.045 / 写入 ¥0.585 | -10% | — | — | — |
| 满血Claude【禁蒸馏】 | claude-opus-4-8 | 输入 ¥4 / 输出 ¥20 / 缓存 ¥0.4 / 写入 ¥5.2 | 0% | — | — | — |
| Channel | Provider model | Current price | Price trend | Uptime | Fake % | Latency |
|---|---|---|---|---|---|---|
| Claude Kiro高缓 | claude-opus-4-7 | 输入 ¥0.45 / 输出 ¥2.25 / 缓存 ¥0.045 / 写入 ¥0.585 | -10% | — | — | — |
| 满血Claude【禁蒸馏】 | claude-opus-4-7 | 输入 ¥4 / 输出 ¥20 / 缓存 ¥0.4 / 写入 ¥5.2 | 0% | — | — | — |
| Channel | Provider model | Current price | Price trend | Uptime | Fake % | Latency |
|---|---|---|---|---|---|---|
| Claude Kiro高缓 | claude-sonnet-5 | 输入 ¥0.27 / 输出 ¥1.35 / 缓存 ¥0.027 / 写入 ¥0.3375 | -10% | — | — | — |
| 满血Claude【禁蒸馏】 | claude-sonnet-5 | 输入 ¥2.4 / 输出 ¥12 / 缓存 ¥0.24 / 写入 ¥3 | 0% | — | — | — |
| Channel | Provider model | Current price | Price trend | Uptime | Fake % | Latency |
|---|---|---|---|---|---|---|
| GPT低价 | gpt-5.6-sol | 输入 ¥0.35 / 输出 ¥2.1 / 缓存 ¥0.035 / 写入 ¥0.4375 | -22% | — | — | — |
| GPT 不降智 | gpt-5.6-sol | 输入 ¥1.25 / 输出 ¥7.5 / 缓存 ¥0.125 / 写入 ¥1.56 | 0% | — | — | — |
| Channel | Provider model | Current price | Price trend | Uptime | Fake % | Latency |
|---|---|---|---|---|---|---|
| GPT 不降智 | gpt-5.6-terra | 输入 ¥0.5 / 输出 ¥3 / 缓存 ¥0.05 / 写入 ¥0.6563 | 0% | — | — | — |
| Channel | Provider model | Current price | Price trend | Uptime | Fake % | Latency |
|---|---|---|---|---|---|---|
| 国模特价 | glm-5.2 | 输入 ¥0.8 / 输出 ¥2.8 / 缓存 ¥0.2 / 写入 ¥0 | 0% | — | — | — |
| 国模主力 | glm-5.2 | 输入 ¥1.2 / 输出 ¥4.2 / 缓存 ¥0.3 / 写入 ¥0 | 0% | — | — | — |
| 国模兜底 | glm-5.2 | 输入 ¥1.6 / 输出 ¥5.6 / 缓存 ¥0.4 / 写入 ¥0 | 0% | — | — | — |
| Channel | Provider model | Current price | Price trend | Uptime | Fake % | Latency |
|---|---|---|---|---|---|---|
| 生图专用 | gpt-image-2 | ¥0.02 / 次 | - | — | — | — |
| Channel | Provider model | Current price | Price trend | Uptime | Fake % | Latency |
|---|---|---|---|---|---|---|
| Claude Kiro高缓 | claude-opus-4-6 | 输入 ¥0.45 / 输出 ¥2.25 / 缓存 ¥0.045 / 写入 ¥0.585 | -10% | — | — | — |
| 满血Claude【禁蒸馏】 | claude-opus-4-6 | 输入 ¥4 / 输出 ¥20 / 缓存 ¥0.4 / 写入 ¥5.2 | 0% | — | — | — |
Uptime is the share of recent probes where the service was available; a higher value generally indicates better stability.
Model consistency uses model pass results to show whether the endpoint behaves like the target model.
Pricing and latency come from public monitoring data and are intended for side-by-side gateway comparison.
Data is provided for reference. Third-party gateways may experience outages or balance loss, so test first and add only a small balance when needed.