CheapRouterAuthenticity, speed and price analysis
(seconds)
coding session
availability
Comparison summary
Gemini 4 Carbon on CheapRouter: a coding session costs ¥12.8, ranking 13 of 15 (median ¥3.08).
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 |
|---|---|---|---|---|---|
| Gemini Vertex | Gemini 4 Carbon | Not probed yet | ¥5.80 | ¥29.00 | Per million tokens |
Cost per coding session
Gemini 4 Carbon · 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
Gemini 4 Carbon · 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 |
|---|---|---|---|---|---|
| Grok 4.6 | Grok Ultra | — | — | — | |
| GPT 5.6 Terra | Codex促销分组-不降智 | — | — | — | |
| GLM 5.3 | 国模按量付费分组 | — | — | — | |
| Claude Opus 5 | Claude促销分组 | — | — | — | |
| GPT 6.2 Sol | Codex促销分组-不降智 | — | — | — | |
| GPT 6.2 Astra | Codex促销分组-不降智 | — | — | — | |
| Claude Fable 5.2 | Claude促销分组 | — | — | — | |
| Claude Sonnet 5.5 | Claude促销分组 | — | — | — | |
| Kimi K3.1 Pro | 国模按量付费分组 | — | — | — | |
| GPT 6.1 Terra | Codex促销分组-不降智 | — | — | — | |
| Gemini 4 Carbon | Gemini Vertex | — | — | — | |
| Gemini 4 Hydrogen | Gemini Vertex | — | — | — | |
| Gemini 4 Argon | Gemini Vertex | — | — | — | |
| GPT 6 Sol | Codex促销分组-不降智 | — | — | — | |
| Kimi K3 | 国模按量付费分组 | — | — | — | |
| Claude Fable 5 | Claude促销分组 | — | — | — | |
| Claude Sonnet 5 | Claude促销分组 | — | — | — | |
| GPT 5.6 Sol | Codex促销分组-不降智 | — | — | — | |
| Claude Opus 4.6 | Claude促销分组 | — | — | — | |
| GLM 5.3 Flash | 国模按量付费分组 | — | — | — | |
| Claude Opus 4.8 | Claude促销分组 | — | — | — | |
| Claude Opus 4.7 | Claude促销分组 | — | — | — | |
| Claude Sonnet 4.6 | Claude促销分组 | — | — | — | |
| Grok 4.5 | Grok Ultra | — | — | — |
About the site
A low-cost, pay-as-you-go gateway for authentic models, with transparent multipliers for Claude, Codex, Gemini, Grok and Chinese models. Offers multi-channel load balancing, ongoing model fingerprint checks and its own Agent client. Register through FolkBench for US$3 in credits.
Service details
- Minimum top-up¥10
- Sign-up bonusNot provided
- Support responseWithin 4 hours on working days
- Payment methodsAlipay, WeChat Pay, Bank card, USDT, Other
- China invoiceSupported
- International invoiceNot supported
- RefundsConditional
- PromotionsNone
FAQ
How much does Gemini 4 Carbon cost on CheapRouter?
Input ¥5.8 and output ¥29 per 1M tokens; at 2M input and 40k output tokens, a coding session costs about ¥12.8, an API rate of 0.22.
How does the cache hit rate affect price?
On CheapRouter, the cache-hit price is 5% of the input price. At an 80% hit rate a coding session costs about ¥3.94, versus about ¥12.8 without cache.Measured hit rates will replace this assumption once available.
Which models does CheapRouter support?
Folkbench currently lists 24: GPT 6 Sol, GPT 6.2 Sol, Claude Sonnet 5.5, GPT 6.2 Astra, Claude Fable 5.2, GPT 6.1 Terra, Claude Fable 5, GPT 5.6 Sol, GPT 5.6 Terra, Claude Opus 5, GLM 5.3 Flash, Claude Opus 4.8, Claude Sonnet 5, Kimi K3.1 Pro, Claude Opus 4.7, Grok 4.6, GLM 5.3, Kimi K3, Claude Opus 4.6, Gemini 4 Carbon, Gemini 4 Argon, Claude Sonnet 4.6, Grok 4.5, Gemini 4 Hydrogen.
What does the data security level mean?
Folkbench rates relays from 0 to 4 based on public policies and verifiable evidence; CheapRouter 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.