WaveAPI
Now routing 100+ models

Every modelone endpoint.

One OpenAI-compatible API in front of every major lab. One key, one bill, automatic failover. Pay only for what you use.

Trusted by 10,000+ developers

Model Catalog

New models, the day they ship.

Priced per million tokens, input / output. The catalog updates live from the gateway.

View all

Quick Start

Swap one line. Keep your SDK.

Point the official client at our base URL. Every request stays OpenAI-compatible — no rewrites, no new SDK.

  1. 1

    Get a key

    One key works across every model and vendor.

  2. 2

    Set the base URL

    Point the official SDK at our gateway — the only change.

  3. 3

    Send a request

    Switch models by name. Billing rolls up in USD.

from openai import OpenAI

client = OpenAI(
    api_key="$QWAVE_API_KEY",
    base_url="https://api.qingwave.com/v1",
)

r = client.chat.completions.create(
    model="gpt-5",
    messages=[{"role": "user", "content": "Hi"}],
)
print(r.choices[0].message.content)

Built to be trusted

No lock-in. No surprises.

OpenAI-compatible

Drop-in for the official SDKs. Change the base URL, keep your code.

Transparent USD billing

Pay as you go, priced per token. No subscription, no minimums.

Data compliant

Clear data handling. Your requests are never used to train models.

Real status, real docs

A live status page and a full reference — never "coming soon" stubs.