Methodology

How We Score Providers

Every provider is scored 0–100 across five equally-weighted dimensions. Scores are calculated manually based on public information, community reports, and direct testing. No score is paid for or influenced by providers. We earn affiliate commissions — disclosed clearly, never at the cost of accuracy.

Scoring Dimensions

Transparency

20 pts

Does the provider have a public about page? Are model names and prices clearly disclosed? Is the pricing model (per-token vs. subscription) unambiguous? Providers that hide credit-to-token conversion rates lose points here.

Example scores

Full pricing page with per-model rates20/20
Subscription tiers without token conversion10/20
No pricing information without signup0/20

Support Quality

20 pts

Can you reach a human when something breaks? Email and ticket-based support score highest. Telegram-only is weak but better than nothing. WeChat-only is a red flag for non-Chinese users.

Example scores

Email + GitHub issue tracker20/20
Telegram community + DM12/20
WeChat QR only5/20

Payment Safety

20 pts

Stripe payments protect users with chargeback rights. Crypto-only or WeChat-only payments offer no recourse if the provider goes offline. We check for which payment processors are accepted.

Example scores

Stripe only20/20
Stripe + crypto16/20
Alipay/WeChat only8/20
Payment method unknown0/20

Community

20 pts

Is there a visible developer community around this service? GitHub repos, Discord servers, real IDE integrations (Claude Code, Cursor, Cline), and social presence all signal that real developers have tried and vouched for the product.

Example scores

GitHub org + Discord + 100K users claimed20/20
Discord community only12/20
No social presence0/20

Longevity

20 pts

How long has the provider been operating? New domains (especially cheap TLDs like .me, .top, .cloud) score lower. Providers with Wayback Machine history, established customer bases, or stated founding dates score higher.

Example scores

Founded 2024 or earlier, verifiable history20/20
Founded 2025, some community evidence12/20
Founded 2026, no archive history5/20

Trust Tiers

Trusted

75–100 pts

Independent verification, strong community, Stripe payments, transparent pricing.

Verify First

50–74 pts

Usable but verify pricing and check for recent community reports before spending.

Caution

25–49 pts

Significant trust gaps. New domain, limited support, or opaque pricing.

Avoid

0–24 pts

Multiple serious red flags. Do not use for sensitive workloads.

Warning Flags

In addition to trust scores, we apply independent warning flags for specific risk factors. These appear on provider cards and in the comparison table.

China-Adjacent Operator

Providers operated from China or using Chinese payment methods (Alipay, WeChat Pay) create data residency concerns. Your API calls — including prompts, code, and documents — may pass through infrastructure subject to Chinese law. This is a risk signal, not an automatic disqualification.

More Expensive Than Official

Some gateways claim to be discount services but price certain models above official Anthropic/OpenAI/Google rates. This can happen due to currency conversion, profit margin on specific models, or simple pricing errors. We flag this prominently.

Custom Model Naming

Some China-adjacent providers use non-standard model names (e.g., gpt-5.2, gpt-5.4) that do not correspond to real OpenAI products. This makes it impossible to verify what model is actually being called.

Credit Systems Without Token Conversion

Subscription services that sell 'credits' without disclosing how many credits equal 1M tokens make price comparison impossible. We flag these providers and exclude them from per-token price comparisons.

Limitations

  • We cannot verify actual API latency or uptime without running sustained load tests — we use self-reported figures where available.
  • Pricing may change at any time. We update weekly but cannot guarantee real-time accuracy.
  • China-adjacent operators may provide excellent service — the flag is a data residency disclosure, not a quality judgment.
  • Trust scores reflect public signals only. A high score does not guarantee a provider won't disappear tomorrow.
  • We have not independently verified SOC2 or other compliance claims. "Claims SOC2" ≠ "is SOC2 certified."