AIWave

A practical DeepSeek API route for US applications.

Point an existing OpenAI-compatible client at AIWave, keep your key server side, and test one small request before you send a production workload. Rates and model IDs are published for verification.

Create your API key Run the quickstart

Start with the request your team needs.

A small smoke test proves the route works. The next request should resemble the workload you actually plan to run.

One representative payload, one saved result
01

Keep USD billing

Use one USD balance and invoice surface instead of changing your application around a new billing format.

Keep USD billing
02

Start with a small request

Use a short prompt to confirm the key, route, model ID, and response before carrying a larger context.

Start with a small request
03

Measure the real call

Save the request status, usage fields, effective group, and rate information with your application logs.

Measure the real call

Check the details that decide a successful call.

The client shape is only the beginning. These checks keep a working request understandable after it leaves your terminal.

Route, account, response, record
01

Exact model ID

Copy the current DeepSeek ID from the model directory. Do not rely on an old alias from a code sample.

02

Authentication

Send Authorization: Bearer with the AIWave key. Keep the key out of browser bundles and repositories.

03

Billing

Check the dated rate and the account group shown in Console before scaling the workload.

Run one small request before you scale it.

Keep the key server side. Copy the request, replace the model only with an ID from the current directory, and inspect the response.

OpenAI-compatible chat completions

Copy-ready request

curl https://aiwave.live/v1/chat/completions \
  -H "Authorization: Bearer $AIWAVE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"deepseek-v4-flash","messages":[{"role":"user","content":"Say hello in one sentence."}],"max_tokens":32}'

Before production

Keep the first result explainable.

Save the request ID, model ID, usage fields, status, and charge with the application record.

  1. Check the key without printing its value.
  2. Use a current model ID from the directory.
  3. Set a client timeout and bound retries.
  4. Compare the response record with the dated rate.

Questions developers ask before integration.

Can I call DeepSeek from the US?

AIWave provides an OpenAI-compatible route for developers who need a documented path and USD billing. Check current model availability before a production call.

Do I need to change my OpenAI client?

Usually not. Set base_url to https://aiwave.live/v1 and use an AIWave key on the server side.

Where can I verify the rate?

Use the dated pricing page and the public pricing JSON, then compare the result with the request record.

Run a request you can inspect afterward.

Keep the request small, confirm the route, then move to the workload that matters.

Open the quickstart Browse model IDs

One more question before you send traffic.

Evidence checked 2026-09-21.

What should I verify before sending a larger DeepSeek request?

Run a small representative request, confirm HTTP 200 and usage fields, then check the dated rate and effective account group. FAQ checked 2026-09-21.