API authenticity checks
Test an API endpoint and model before relying on it. Hvoy AI checks connectivity, response behavior, latency, and model consistency signals.
Hvoy AI tested rankings
Tested AI Gateway Recommendations
Compare tested Claude, GPT, and Gemini gateways by uptime, model consistency, latency, pricing, and sample size, then open the full leaderboard for detailed channel data.
Measured recommendations
Rankings use recent public probes and pricing data. Third-party gateways may change, so test first and start with a small balance.
AI gateway ranking snapshot
Data updated: Aug 8, 2026
GPT 5.6 Sol
GPT 5.5
| Rank | Recommended gateway | Uptime | Latency | Input price |
|---|---|---|---|---|
| 1 | CUN.aicun.ai | 100% | 12.8s | ¥3.65/M tokens |
| 2 | hao.aihao.ai | 100% | 10.7s | ¥5.1/M tokens |
| 3 | OfficesAIofficesai.top | 81% | 13.1s | ¥0.825/M tokens |
| 4 | tallius.aiapi.tallius.ai | 96% | 15.2s | ¥0.945/M tokens |
| 5 | DawCodedawclaudecode.com | 100% | 9.1s | ¥2/M tokens |
| 6 | zTokenhttps://ai.ztoken.dev | 96% | 11.3s | ¥1.5/M tokens |
Opus 5
| Rank | Recommended gateway | Uptime | Latency | Input price |
|---|---|---|---|---|
| 1 | CUN.aicun.ai | 100% | 5.5s | ¥3.65/M tokens |
| 2 | hao.aihao.ai | 100% | 5.3s | ¥5.1/M tokens |
| 3 | Monikeraimoniker.top | 96% | 4.7s | ¥6/M tokens |
| 4 | UU APIuuapi.net | 100% | 4.0s | ¥4/M tokens |
| 5 | relaycorerelaycore.cn | 100% | 6.4s | ¥5/M tokens |
| 6 | ClaudeCNclaudecn.top | 96% | 5.9s | ¥8/M tokens |
Fable 5
GPT Image 2
| Rank | Recommended gateway | Uptime | Latency | Input price |
|---|---|---|---|---|
| 1 | CUN.aicun.ai | 85% | 35.1s | ¥5.84/M tokens |
| 3 | Lumin AIapi.luminai.cc | 94% | 25.4s | ¥0.15/M tokens |
| 4 | PackyCodepackyapi.ai | 100% | 32.9s | ¥0.08/M tokens |
| 5 | SadAIsadai.cc | 100% | 25.0s | ¥0.041/M tokens |
| 6 | UU APIuuapi.net | 93% | 34.9s | — |
| 7 | JBB金贝贝jbbt.cc | 92% | 34.0s | ¥0.06/M tokens |
Opus 4.8
| Rank | Recommended gateway | Uptime | Latency | Input price |
|---|---|---|---|---|
| 1 | CUN.aicun.ai | 96% | 4.1s | ¥3.65/M tokens |
| 2 | hao.aihao.ai | 100% | 3.5s | ¥5.1/M tokens |
| 3 | tallius.aiapi.tallius.ai | 100% | 5.7s | ¥5.25/M tokens |
| 4 | ZhonZhuan.aizhonzhuan.ai | 96% | 4.5s | — |
| 5 | DawCodedawclaudecode.com | 96% | 7.1s | ¥10/M tokens |
| 6 | UoCodeuocode.com | 100% | 4.7s | — |
GPT 5.6 Terra
DeepSeek V4 Flash
| Rank | Recommended gateway | Uptime | Latency | Input price |
|---|---|---|---|---|
| 1 | DuiAPIduiapi.com | 100% | 2.5s | ¥0.7/M tokens |
| 2 | CUN.aicun.ai | 96% | 2.7s | ¥0.1/M tokens |
| 3 | RunAPIrunapi.co | 100% | 2.9s | ¥0.776/M tokens |
| 4 | MoreCodeapi.1314mc.net | 96% | 5.9s | — |
| 5 | UU APIuuapi.net | 100% | 2.9s | ¥0.65/M tokens |
| 6 | AutoRouterautorouter.net | 100% | 4.8s | ¥0.6/M tokens |
Kimi K3
| Rank | Recommended gateway | Uptime | Latency | Input price |
|---|---|---|---|---|
| 1 | DuiAPIduiapi.com | 96% | 5.1s | ¥20/M tokens |
| 2 | CUN.aicun.ai | 100% | 6.0s | ¥18.396/M tokens |
| 3 | Frontier Intelligenceapi.frontier-intelligence.tech | 96% | 5.2s | ¥30/M tokens |
| 4 | 8sToken8stoken.com | 85% | 4.0s | ¥30/M tokens |
| 5 | 悟空APIhttps://wkapi.pro | 100% | 6.0s | ¥0.8/M tokens |
| 6 | Modelflarehttps://origin.modelflare.dev | 100% | 6.5s | ¥3/M tokens |
Grok 4.5
GLM 5.2
| Rank | Recommended gateway | Uptime | Latency | Input price |
|---|---|---|---|---|
| 1 | DuiAPIduiapi.com | 100% | 4.0s | ¥4.8/M tokens |
| 2 | CUN.aicun.ai | 100% | 3.8s | ¥4.38/M tokens |
| 3 | routescoperoutescope.ai | 100% | 3.6s | — |
| 4 | FluxionAIfluxionai.space | 100% | 3.5s | ¥3.2/M tokens |
| 5 | UU APIuuapi.net | 100% | 4.3s | ¥5.2/M tokens |
| 6 | AutoRouterautorouter.net | 100% | 5.4s | ¥4.8/M tokens |
Opus 4.6
| Rank | Recommended gateway | Uptime | Latency | Input price |
|---|---|---|---|---|
| 1 | CUN.aicun.ai | 100% | 5.3s | ¥3.65/M tokens |
| 2 | hao.aihao.ai | 93% | 5.8s | ¥5.1/M tokens |
| 3 | OfficesAIofficesai.top | 100% | 4.8s | ¥6.6/M tokens |
| 4 | ZhonZhuan.aizhonzhuan.ai | 96% | 6.2s | — |
| 5 | tallius.aiapi.tallius.ai | 85% | 9.5s | ¥5.25/M tokens |
| 6 | Lumin AIapi.luminai.cc | 96% | 2.7s | — |
Sonnet 5
Opus 4.7
| Rank | Recommended gateway | Uptime | Latency | Input price |
|---|---|---|---|---|
| 1 | CUN.aicun.ai | 100% | 5.4s | ¥3.65/M tokens |
| 2 | hao.aihao.ai | 96% | 4.2s | ¥5.1/M tokens |
| 3 | OfficesAIofficesai.top | 100% | 4.9s | ¥6.6/M tokens |
| 4 | tallius.aiapi.tallius.ai | 100% | 4.7s | ¥5.25/M tokens |
| 5 | ZhonZhuan.aizhonzhuan.ai | 89% | 4.5s | — |
| 6 | Lumin AIapi.luminai.cc | 100% | 3.6s | ¥7.5/M tokens |
Common questions
What does the API check cover?
It checks endpoint connectivity, response behavior, latency, and model consistency signals.
How are gateway recommendations produced?
Recommendations combine recent public probes, pricing data, model coverage, and available site information.
Are third-party gateways risk-free?
No. Pricing, billing, and stability can change, so test with a small balance first.