Quickstart

Kiln serves an OpenAI-compatible API. Get a response from qwen3-32b in three steps: create an account, create an API key, and point an OpenAI client at the Kiln endpoint.

1. Sign up

Create an account on the sign-up page with an email address or with GitHub, Google or Naver. A personal organization is created for you; keys, usage and credits belong to it.

Verify your email address. Trial credit is granted once, after verification, and expires 30 days later. The credits page in the console shows the amount and the remaining balance.
During the open beta an operator may need to approve your account before the console opens. The pending page shows both checks and refreshes on its own.

2. Create an API key

In the console → API keys, an organization admin creates a key for a project. Keys start with sk-bk- and the full value is shown only once, at creation, so copy it then. Key creation is refused until you verify your email and your account is approved.

Tip

Keep your API key secret. Never commit it to source control or ship it in client-side code. Load it from an environment variable (e.g. KILN_API_KEY) and rotate it if it leaks.

3. Send your first request

Use an OpenAI-compatible client. The official OpenAI SDKs are the quickest way to start:

bash

pip install openai

bash

npm install openai

Set your client's base_url to https://api.bricksum.com/v1. Request and response shapes follow the OpenAI API. Some parameters are not supported yet; see Known limitations in the API reference.

base_url

base_url = "https://api.bricksum.com/v1"

Send a chat/completions request. The response shape matches the OpenAI API, including the usage field used for billing.

bash

curl https://api.bricksum.com/v1/chat/completions \
  -H "Authorization: Bearer sk-bk-..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen3-32b",
    "messages": [
      {"role": "user", "content": "Hello!"}
    ]
  }'

python

from openai import OpenAI

client = OpenAI(
    base_url="https://api.bricksum.com/v1",
    api_key="sk-bk-...",  # your Kiln API key
)

resp = client.chat.completions.create(
    model="qwen3-32b",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(resp.choices[0].message.content)

javascript

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.bricksum.com/v1",
  apiKey: process.env.KILN_API_KEY, // sk-bk-...
});

const resp = await client.chat.completions.create({
  model: "qwen3-32b",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(resp.choices[0].message.content);

Streaming

Set stream to true to receive tokens as server-sent events (SSE) as they are generated.

bash

curl https://api.bricksum.com/v1/chat/completions \
  -H "Authorization: Bearer sk-bk-..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen3-32b",
    "messages": [{"role": "user", "content": "Hello!"}],
    "stream": true
  }'

python

from openai import OpenAI

client = OpenAI(
    base_url="https://api.bricksum.com/v1",
    api_key="sk-bk-...",
)

stream = client.chat.completions.create(
    model="qwen3-32b",
    messages=[{"role": "user", "content": "Hello!"}],
    stream=True,
)
for chunk in stream:
    print(chunk.choices[0].delta.content or "", end="")

javascript

import OpenAI from "openai";

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

const stream = await client.chat.completions.create({
  model: "qwen3-32b",
  messages: [{ role: "user", content: "Hello!" }],
  stream: true,
});
for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0].delta.content ?? "");
}

Next steps

Model catalog — compare current pricing, measured throughput over the last 24 hours and 90-day uptime for each model, then put the model ID in the model field
API Reference — supported endpoints, request fields and known limitations
Errors — what 401 (bad key), 402 (no credit) and 429 (rate limit) mean, and how to retry
Rate limits — rate limits (default 60 RPM) and retry guidance
Console → Tools — copy the setup snippet for Claude Code, Codex CLI, OpenCode or Cursor