// Docs · Quickstart · 5 min

Five minutes to your first half-price call.

SLASHED is an OpenAI-compatible gateway. If your code already speaks OpenAI's API, this is a one-string swap. If not, it's three lines.

// What you'll do 1. Grab a key from the dashboard. 2. Set base_url to https://api.slashed.pro/v1. 3. Make a request with the official OpenAI SDK or plain cURL. That is the whole tutorial.

01Get a key

Sign in at dashboard.slashed.pro and create a new key from the Keys tab. Keys are prefixed sl- (eight chars of identifier, then secret material). Copy it once — the secret half is shown a single time at creation.

You can name the key, scope it to specific model IDs, and set a monthly spend ceiling. Defaults: all models, no ceiling. See /docs/authentication for the full surface.

02Set your base URL

The only line of your existing code that changes is the base URL. Everything else — request shape, response shape, streaming, tool-use, JSON mode — follows the OpenAI v1 spec as the upstream provider supports it.

Field	Value
`base_url`	`https://api.slashed.pro/v1`
`Authorization`	`Bearer sl-...`
`model`	Any ID from /docs/models

03Make a request

Pick a language. Each pane is a complete, self-contained example.

cURL

Python

Node

# Documented: set your sl- key from https://dashboard.slashed.pro
curl https://api.slashed.pro/v1/chat/completions \
  -H "Authorization: Bearer $SLASHED_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.4-mini",
    "messages": [
      {"role": "user", "content": "Reply with the word: ready"}
    ]
  }'

# pip install openai — the official client works unchanged
from openai import OpenAI

client = OpenAI(
    base_url="https://api.slashed.pro/v1",
    api_key="sl-...",
)

resp = client.chat.completions.create(
    model="gpt-5.4-mini",
    messages=[{"role": "user", "content": "Reply with the word: ready"}],
)
print(resp.choices[0].message.content)

// npm i openai — the official client works unchanged
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.slashed.pro/v1",
  apiKey: "sl-...",
});

const resp = await client.chat.completions.create({
  model: "gpt-5.4-mini",
  messages: [{ role: "user", content: "Reply with the word: ready" }],
});
console.log(resp.choices[0].message.content);

04Verify the model list

The list of models you can request is canonical at /v1/models. No auth required — it's a public document of what the gateway can route to.

curl https://api.slashed.pro/v1/models

Returns an OpenAI-spec list with id, object, created, and owned_by for each entry. See /docs/models for the documented entries and the capability matrix.

05Where to go next

Coming from Quatarly? See /docs/migrate. Step-by-step key and base_url swap.
Need streaming, tool-use, vision? See /docs/endpoints.
Hit an error? Every error message links to a row in /docs/errors.
Building an integration? The polished single-page version of this is at /quickstart.

// Note on capabilities Streaming, vision, tool-use, and JSON mode are documented as supported per the upstream model's spec. The gateway proxies the upstream contract — if the upstream model supports the field, the gateway forwards it. We do not synthesize features the upstream provider does not expose.

Authentication