Which models are supported?

We support OpenAI, Anthropic, Google, DeepSeek, Qwen, and dozens of other providers. More models are being added regularly.

How does billing work?

We use a pay-as-you-go model. You are only charged for the tokens you actually use. All pricing is transparent with no hidden fees.

Is there a free trial?

Yes! New users receive $1 in free credits upon registration. You can fully test our service before making any payment.

Is the API compatible with OpenAI format?

Yes, fully compatible. You can use any OpenAI SDK or client library by simply changing the Base URL and API Key.

How do you ensure service reliability?

We use multi-node deployment with health checks and intelligent routing to ensure high availability.

How do I get started?

Register for a free account, create an API key from the dashboard, and start making requests. You get $1 free credit to test everything.

Can I use my existing OpenAI SDK integration?

Yes. Just update the Base URL to our endpoint and replace your API key. No code changes needed — our API is fully OpenAI-compatible.

What happens if a provider goes down?

Our intelligent routing engine automatically detects provider failures and retries or falls back to an alternative provider, ensuring minimal disruption.

How do you compare to using provider APIs directly?

We simplify multi-provider management with one integration, unified billing, automatic failover, and detailed usage analytics — no need to manage multiple API keys and billing accounts.

Do you support streaming responses?

Yes, we fully support Server-Sent Events (SSE) streaming for all major providers, matching the OpenAI streaming format.

Can I set usage limits per API key?

Yes, each API key can have configurable rate limits, total quota, and remaining quota controls for granular access management.

What payment methods do you accept?

We accept credit/debit cards via Stripe, Alipay and WeChat Pay via Epay, and redemption code top-ups.

Do you offer custom enterprise pricing?

Yes, for high-volume users and enterprises, we offer custom pricing and dedicated support. Contact our sales team for details.

All API traffic is encrypted in transit via TLS 1.3. Your API keys are stored using SHA-256 hashing and provider credentials are encrypted at rest with AES-256-GCM.

Do you support embedding models?

Yes, we support embedding APIs from OpenAI, Google, Cohere, Jina, and other major providers through our unified embedding endpoint.

How can I monitor my API usage?

The dashboard provides real-time usage analytics including token counts, cost breakdowns, request volume, and latency metrics per model and per API key.

Connect every model. One API.

One gateway.Every model.

Build faster with unified billing and real-time cost control across dozens of model providers. One API, zero lock-in.

Select client · Copy · Run

copy & run

from anthropic import Anthropic

client = Anthropic(

api_key="sk-...",

base_url=https://gateway.hkting.com

)

resp = client.messages.create(

model="claude-sonnet-5",

max_tokens=1024,

messages=[{"role": "user", "content": "hello"}]

)

<15ms

Gateway overhead

99.9%

Uptime SLA

7×24

Stable relay

$0.55

/1K tokens from