超稳API / Stable APIAuthenticity, speed and price analysis
About the site
超稳API / Stable API 是 OpenAI 兼容的多模型 API 网关/聚合服务,按量付费额度不过期(以公开定价页为准)。支持定价页列出的 GPT、Claude、Gemini、DeepSeek 等模型,以及文本、图像和音视频能力。官网:https://apiniuma.com ;定价:https://apiniuma.com/pricing ;文档:https://docs.apiniuma.com/
Service details
- Minimum top-up$1
- Sign-up bonusNot provided
- Support response新加坡时间周一至周五 10:00-17:30(以官网公示客服时间为准)
- Payment methodsAlipay, WeChat Pay, Other
- China invoiceNot verified
- International invoiceNot verified
- RefundsNot verified
- PromotionsNot provided
(seconds)
coding session
availability
Comparison summary
Qwen 3.5 Plus on 超稳API / Stable API: a coding session costs ¥1.34, ranking 2 of 3 (median ¥1.34).
Authenticity
Model consistency
Output distribution for the same requests on this relay and on the official API · higher is better
Speed
Measured by time to first token (seconds) and output speed (tokens / second)
Time to first token
Seconds until the first token · lower is better
Price
Groups and models
| Group | Model | Availability | Input price | Output price | Pricing unit |
|---|---|---|---|---|---|
| Alibaba-2 | Qwen 3.5 Plus | Not probed yet | ¥0.60 | ¥3.60 | Per million tokens |
| Alibaba-3 | Qwen 3.5 Plus | Not probed yet | ¥0.88 | ¥5.28 | Per million tokens |
| Alibaba-1 | Qwen 3.5 Plus | Not probed yet | ¥0.40 | ¥2.40 | Per million tokens |
Cost per coding session
Qwen 3.5 Plus · 2M input and 40k output tokens (¥) · split by token type · lower is better
Prices: cache hit, input and output
¥ / 1M tokens
Cache
Coding requests repeat much of their context, so the cache hit rate largely decides what you actually pay. Drag the hit rate and the costs and ranking below are reweighted with cache prices.
Cache-weighted session cost
Qwen 3.5 Plus · of 2M input tokens, the hit share is billed at the cache price and the rest at the input price, plus 40k output (¥) · lower is better
Stability
Availability
Request success rate, last 24 hours · higher is better
Status of each model, last 24 hours
One cell per hour · green ok, yellow delayed, red failed
| Model | Group | Availability | Last 24 hours | First token | Samples |
|---|---|---|---|---|---|
| Gemini 2.5 Flash Lite | Vertex-Gemini-2 | — | — | — | |
| GPT 5.3 Codex | Openai-Gpt-1 | — | — | — | |
| DeepSeek V4 Flash 0731 | Self-Deployed-1 | — | — | — | |
| o3 | Azure-Gpt-3 | — | — | — | |
| Grok Build 0.1 | Cli-Grok-2 | — | — | — | |
| GPT 5 Mini | Azure-Gpt-3 | — | — | — | |
| Qwen 3.8 Flash | Alibaba-2 | — | — | — | |
| Qwen 3.7 Max | Alibaba-2 | — | — | — | |
| GPT 5.4 | Azure-Gpt-3 | — | — | — | |
| GPT 5.2 | Azure-Gpt-3 | — | — | — | |
| GPT 5.4 Mini | Azure-Gpt-3 | — | — | — | |
| GPT 5.6 Luna | Azure-Gpt-3 | — | — | — | |
| GPT-4o Mini | Azure-Gpt-3 | — | — | — | |
| Gemini 2.5 Flash | Vertex-Gemini-2 | — | — | — | |
| GPT-4o | Azure-Gpt-3 | — | — | — | |
| GPT 4.1 | Azure-Gpt-3 | — | — | — | |
| GPT 4.1 Mini | Azure-Gpt-3 | — | — | — | |
| Gemini 3.5 Flash | Aistudio-Gemini-2 | — | — | — | |
| GPT 5 | Azure-Gpt-3 | — | — | — | |
| GPT 5 Nano | Azure-Gpt-3 | — | — | — | |
| Gemini 4 Argon | Aistudio-Gemini-2 | — | — | — | |
| GPT 5.5 | Azure-Gpt-3 | — | — | — | |
| Mimo V2.5 Pro | Xiaomi-1 | — | — | — | |
| Mimo V2.5 | Xiaomi-1 | — | — | — | |
| Hy 4 Preview | hunyuan-1 | — | — | — | |
| MiniMax M2.5 | Self-Deployed-3 | — | — | — | |
| MiniMax M2.7 | Hailuo-3 | — | — | — | |
| MiniMax M3 | Hailuo-3 | — | — | — | |
| Kimi K2.7 Code | Self-Deployed-1 | — | — | — | |
| Kimi K2.6 | Self-Deployed-1 | — | — | — | |
| Kimi K3 | Kimi-1 | — | — | — | |
| GLM 5.1 | Alibaba-2 | — | — | — | |
| Qwen 3.6 Plus | Alibaba-2 | — | — | — | |
| DeepSeek V4 Pro | Self-Deployed-1 | — | — | — | |
| GLM 5 | Self-Deployed-3 | — | — | — | |
| Qwen 3.5 Plus | Alibaba-2 | — | — | — | |
| Qwen 3.7 Plus | Alibaba-2 | — | — | — | |
| Qwen 3.8 Max | Alibaba-2 | — | — | — | |
| GLM 5.3 Flash | Self-Deployed-1 | — | — | — | |
| Claude Fable 5 | AWS-Claude-3 | — | — | — | |
| GPT 5.6 Sol | Azure-Gpt-3 | — | — | — | |
| Claude Opus 5 | Azure-Claude-1 | — | — | — | |
| Claude Sonnet 5 | Azure-Claude-1 | — | — | — | |
| Claude Opus 4.8 | Azure-Claude-1 | — | — | — | |
| GPT 5.6 Terra | Openai-Gpt-1 | — | — | — | |
| Claude Opus 4.7 | AWS-Claude-3 | — | — | — | |
| GLM 5.3 | Alibaba-3 | — | — | — | |
| Grok 4.6 | Cli-Grok-2 | — | — | — | |
| Gemini 3.8 Flash | Aistudio-Gemini-2 | — | — | — | |
| GLM 5.2 | Alibaba-2 | — | — | — | |
| Claude Sonnet 4.6 | AWS-Claude-3 | — | — | — | |
| Claude Opus 4.6 | AWS-Claude-3 | — | — | — | |
| Grok 4.5 | Cli-Grok-2 | — | — | — | |
| Gemini 3.6 Flash | Aistudio-Gemini-2 | — | — | — | |
| Gemini 3.7 Flash | Aistudio-Gemini-2 | — | — | — | |
| Gemini 3.1 Pro Preview | Aistudio-Gemini-2 | — | — | — | |
| Gemini 3 Flash Preview | Aistudio-Gemini-2 | — | — | — |
FAQ
How much does Qwen 3.5 Plus cost on 超稳API / Stable API?
Input ¥0.6 and output ¥3.6 per 1M tokens; at 2M input and 40k output tokens, a coding session costs about ¥1.34, an API rate of 0.22.
How does the cache hit rate affect price?
超稳API / Stable API has not published a cache price for this route, so cache hits are billed at the input price.Measured hit rates will replace this assumption once available.
Which models does 超稳API / Stable API support?
Folkbench currently lists 57: GLM 5.3 Flash, Claude Fable 5, Claude Opus 5, GPT 5.6 Sol, Claude Sonnet 5, Claude Opus 4.8, Kimi K3, GPT 5.6 Terra, GLM 5.3, Gemini 4 Argon, Gemini 3.8 Flash, Grok 4.6, DeepSeek V4 Pro, Claude Opus 4.7, Claude Sonnet 4.6, Claude Opus 4.6, Grok 4.5, Qwen 3.8 Max, GLM 5.2, Qwen 3.8 Flash, GPT 5.6 Luna, MiniMax M3, Gemini 3.6 Flash, Gemini 3.7 Flash, Gemini 3.1 Pro Preview, Gemini 3.5 Flash, Hy 4 Preview, Qwen 3.7 Max, GPT 5.5, Mimo V2.5 Pro, Kimi K2.7 Code, Kimi K2.6, DeepSeek V4 Flash 0731, Gemini 2.5 Flash, Mimo V2.5, MiniMax M2.7, GLM 5.1, Qwen 3.7 Plus, Gemini 3 Flash Preview, GPT-4o Mini, GPT 5.4, GPT 5.2, GPT 5.4 Mini, GPT-4o, MiniMax M2.5, Qwen 3.6 Plus, GLM 5, Qwen 3.5 Plus, Gemini 2.5 Flash Lite, GPT 5.3 Codex, Grok Build 0.1, GPT 4.1, GPT 4.1 Mini, GPT 5, GPT 5 Nano, o3, GPT 5 Mini.
What does the data security level mean?
Folkbench rates relays from 0 to 4 based on public policies and verifiable evidence; 超稳API / Stable API is currently level 0. The check covers leaks in responses and error messages only, and does not prove data retention or training policies.
How often is this data updated?
Availability and time to first token use a 24-hour window, and this page syncs live data every 15 minutes.