API Reference

Base URL

https://adzukiai.com/v1

Authentication

Send Authorization: Bearer sk-... on every request. Keys are created and revoked in the console, and are displayed in full exactly once.

Billing

Prepaid balance, charged per token. See Models & pricing for the per-million rates.

curl

curl https://adzukiai.com/v1/chat/completions \
  -H "Authorization: Bearer $ADZUKI_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v4.1-flash",
    "messages": [
      {"role": "user", "content": "Explain BBR in two sentences."}
    ]
  }'

python — openai sdk

from openai import OpenAI

client = OpenAI(
    base_url="https://adzukiai.com/v1",
    api_key="sk-...",          # your Adzuki AI key
)

resp = client.chat.completions.create(
    model="deepseek-v4.1-flash",
    messages=[{"role": "user", "content": "Explain BBR in two sentences."}],
)
print(resp.choices[0].message.content)

node — openai sdk

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://adzukiai.com/v1",
  apiKey: process.env.ADZUKI_KEY,
});

const resp = await client.chat.completions.create({
  model: "deepseek-v4.1-flash",
  messages: [{ role: "user", content: "Explain BBR in two sentences." }],
});
console.log(resp.choices[0].message.content);

Endpoints

Method	Path	Purpose
`POST`	`/v1/chat/completions`	Chat completion, streaming or buffered
`GET`	`/v1/models`	List the models your key is allowed to call

Request body

Field	Type	Notes
`model`	string	Required. One of the model ids below.
`messages`	array	Required. `role` is `system`, `user` or `assistant`.
`stream`	boolean	Default `false`. Set to `true` for server-sent events.
`max_tokens`	integer	Upper bound on generated tokens. Deep reasoning models may spend part of it on internal reasoning.
`temperature`	number	0–2. Provider defaults apply if omitted.
`top_p`	number	Nucleus sampling. Use it or `temperature`, not both.
`stop`	string / array	Up to four stop sequences.

Streaming

With stream: true the response is text/event-stream. Each event carries a delta; the stream terminates with a literal [DONE] line. Disable any client-side response buffering or you will see the whole answer arrive at once.

curl

curl -N https://adzukiai.com/v1/chat/completions \
  -H "Authorization: Bearer $ADZUKI_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"glm-5.3","stream":true,"messages":[{"role":"user","content":"hi"}]}'

# each event is a line: data: {...}\n\n
# the stream ends with: data: [DONE]

Models

Seven models, four vendors. USD per 1 million tokens; the full table with cached input rates is on Models & pricing.

Model id	Vendor	Input	Output
`doubao-seed-2.0-lite`	ByteDance Doubao	0.17	1.03
`glm-5.3-flash`	Zhipu AI	0.23	0.80
`deepseek-v4.1-flash`	DeepSeek	0.57	2.29
`doubao-seed-2.1-pro`	ByteDance Doubao	1.71	8.57
`glm-5.3`	Zhipu AI	2.29	8.00
`deepseek-v4-pro`	DeepSeek	2.57	7.71
`kimi-k3`	Moonshot AI	5.71	28.57

Errors

Errors use the OpenAI error envelope: {"error": {"message": ..., "type": ...}}. Retry on 429 and 5xx with exponential backoff.

Status	Meaning	What to do
`401`	Key missing, revoked or malformed	Check the `Authorization` header
`402`	Balance exhausted	Top up in the console
`403`	Model not permitted for this key	Remove the model restriction on the key, or use a permitted model
`404`	Unknown model id	Compare against `/v1/models`
`429`	Rate limited	Back off and retry; limits are per key and per IP
`5xx`	Upstream provider failure	Retry with backoff. Failed calls are not charged

Compliance and data handling

!DeepSeek models are restricted for U.S. federal use

Under the FY2026 National Defense Authorization Act (§1532 and §6604) and the U.S. Department of Homeland Security directive issued in October 2026, U.S. federal civilian agencies, the Department of Defense, the Intelligence Community and their contractors may not procure or use DeepSeek technology — including through third-party services such as this one, and regardless of whether the model is self-hosted. Several states, including New York, Virginia, Tennessee and Texas, have imposed comparable restrictions on government devices and networks.

If you are a U.S. federal agency, a federal contractor, or a state or local government entity, do not use the DeepSeek models offered here. Other models in the catalogue are not covered by this restriction. This notice is provided for compliance convenience and is not legal advice; you remain responsible for your own regulatory position.

iWhere your data is processed

Every model in this catalogue is operated by a provider whose inference infrastructure is located in mainland China. When you call a model, the content of your request is transmitted to that provider and processed there, and the response is returned to you. Account and billing data are stored in the United States and the European Union.

We do not use your prompts or outputs to train models. If your workload cannot leave a particular jurisdiction, contact support@adzukiai.com before sending production traffic.