Data 2026-09-04 18:35 · USD/CNY 6.74
中文← Sinan Lab › All relay sites › apibest.ai
apibest.aiAPI BEST
Ultra-lowVisit site →
Price profile
Ultra-low
Median 11%
24h reachable
100%
Latency p50 1428ms · 42 probes · US-West probe node
Models sold
14
Within range 0 · below cost floor 6
Top-up rate
¥7.2 per $1 · Stripe 7.2
Panel "price" field: CNY charged per $1 of nominal credit
Consistency probe
—
Not probed yet; requires a key registered at this site
Login methods
Email, GitHub
No CAPTCHA
Panel
one-api
First indexed 2026-09-02 · source directory:howtok
Models it sells and effective prices
Reference = lowest of official and public market; ratio = effective ÷ reference.
| Model | Effective | Reference price | Effective price as % of reference | How to read | Evidence |
|---|---|---|---|---|---|
| GPT-5.4 gpt-5.4 | 1.60$/M output · fetched 09-02 13:22 | 15.00 | 11% | Mathematically unsustainable | |
| GPT-5.5 gpt-5.5 | 3.21$/M output · fetched 09-02 13:22 | 30.00 | 11% | Mathematically unsustainable | |
| Grok 4.5 grok-4.5 | 0.641$/M output · fetched 09-02 13:22 | 6.00 | 11% | Mathematically unsustainable | |
| GPT-5.4 mini gpt-5.4-mini | 0.481$/M output · fetched 09-02 13:22 | 4.50 | 11% | Mathematically unsustainable | |
| Grok 4.6 grok-4.6 | 0.641$/M output · fetched 09-02 13:22 | 6.00 | 11% | Mathematically unsustainable | |
| GPT-5.6 Luna gpt-5.6-luna | 0.171$/M output · fetched 09-02 13:22 | 1.20 | 14% | Mathematically unsustainable | |
| GPT-5.6 Sol gpt-5.6-sol | 3.42$/M output · fetched 09-02 13:22 | 10.00 | 34% | Below common bulk discounts | |
| GPT-5.6 Terra gpt-5.6-terra | 64.12$/M output · fetched 09-02 13:22 | 12.00 | 534% | Well above public price | |
| Grok 4.20.0309 non reasoning grok-4.20-0309-non-reasoning | 0.641$/M output · fetched 09-02 13:22 | — | — | No reference price | |
| Grok 4.20 fast grok-4.20-fast | 0.053$/M output · fetched 09-02 13:22 | — | — | No reference price | |
| Grok imagine image lite grok-imagine-image-lite | 0.002$/call · fetched 09-02 13:22 | — | — | No reference price | |
| GPT-image 2 gpt-image-2 | 0.006$/call · fetched 09-02 13:22 | — | — | No reference price | |
| Grok 4.20 multi agent xhigh grok-4.20-multi-agent-xhigh | 0.641$/M output · fetched 09-02 13:22 | — | — | No reference price | |
| codex auto review codex-auto-review | 1.60$/M output · fetched 09-02 13:22 | — | — | No reference price |
This is an arithmetic ratio. It is not an accusation against the vendor, and a lower public source not yet indexed here may exist.
Featured models it sellsGPT-5.6 LunaGrok 4.6GPT-5.5GPT-5.6 TerraGPT-5.6 SolGPT-5.4GPT-5.4 miniGrok 4.5
Embeddable measurement badge for site owners
Shows measured values only: 24h reachability, price profile and date, refreshed daily; clicking returns to this page. No opinions.
<a href="https://compute.sinanlab.com/en/s/apibest.ai"><img src="https://compute.sinanlab.com/badge/apibest.ai.svg" alt="Sinan Compute 实测" width="340" height="64"></a>Want to verify it serves the real model? Run the open-source protocol-consistency check (our probe script or Veridrop) with your own key for that site. We only measure price and reachability; we don't judge for you. Outbound links carry no referral parameters; we only count clicks.