Can I smoke test an OpenAI-compatible API without writing code?

Yes. Import the TKEN Postman collection or open the Bruno collection, set the base URL to https://www.tken.shop/v1, add an API key in a local or secret variable, run /models, then send one small /chat/completions request.

Which endpoint should I test first?

Start with GET /models because it checks the base URL, API key, account access, and model discovery without sending prompt content. Choose a model ID returned by /models before testing /chat/completions.

Where should API keys be stored in API clients?

Use local, secret, or environment-scoped variables and avoid committing keys, sharing synced secrets, pasting them into screenshots, or storing them in public collection files.

API-client smoke tests

Smoke test an OpenAI-compatible API with Postman or Bruno

Q: Is TKEN affiliated with Postman, Bruno, OpenAI, or model providers?

No. TKEN is an independent third-party OpenAI-compatible API gateway and is not officially affiliated with Postman, Bruno, OpenAI, or other model providers.

Validate TKEN from a familiar API client before writing app code. Import the collection, keep secrets local, verify /models, choose a returned model ID, then send one small /chat/completions request against the OpenAI-compatible base URL.

Get API key Open GitHub guide

No-code preflight Postman or Bruno

base_url = https://www.tken.shop/v1
api_key = keep in a local or secret variable

1. Import the TKEN collection
2. Run GET /models
3. Copy a returned model ID
4. Run POST /chat/completions
5. Save only redacted evidence

Base URL: https://www.tken.shop/v1

Postman collection and Bruno collection ready

Verify /models before /chat/completions

What this proves

A no-code client smoke test separates gateway configuration from SDK, framework, or app bugs. Use it before connecting Cursor, Continue, Open WebUI, LiteLLM, Vercel AI SDK, LangChain, MCP hosts, or production services.

Reachability

The /v1 gateway route responds

Authentication

The API key is accepted by TKEN

Discovery

GET /models returns available IDs

Chat path

POST /chat/completions succeeds

Secret boundary

Keys stay out of committed files

Next step

Move the same config into SDK code

Postman path

1. Import the collection

Use the TKEN Postman collection from the GitHub examples repo. It keeps the first run focused on model discovery and one chat completion instead of a large automated suite.

2. Set scoped variables

Set base_url to https://www.tken.shop/v1 and store the key in a local, sensitive, or vault-backed value according to your team workflow. Avoid saving real keys into shared collection JSON.

3. Run /models first

Confirm that /models returns IDs for the current account and route. Copy one returned model ID into the chat request instead of assuming a model name from another account or provider page.

4. Send one small chat request

Use a non-sensitive prompt, disable broad automation, and inspect status code, response shape, latency, and a short redacted preview before moving the config into application code.

Bruno path

Bruno works well for Git-tracked API-client collections because request files can stay readable while secrets remain local or secret-scoped. Use the same route order: /models, then one chat completion.

Collection

Open the Bruno folder from GitHub

Environment

Set base_url and model_id per workspace

Secret

Use a local or secret variable for api_key

First request

GET /models

Second request

POST /chat/completions

Evidence

Save status and redacted response only

Safe evidence checklist

Save

Base URL, env var names, status codes, model count, selected model ID, latency, timeout, and a short redacted response preview.

Do not save

API keys, account identifiers, order IDs, usage records, provider account screens, private prompts, customer data, or full response bodies that may contain sensitive content.

When it fails

Classify the failure before changing code: missing /v1, invalid key, empty model list, unavailable model ID, 429, timeout, 5xx, or response-shape mismatch.

TKEN is an independent third-party OpenAI-compatible API gateway and is not officially affiliated with Postman, Bruno, OpenAI, or other model providers. Model availability, pricing, context limits, streaming behavior, tool support, and rate limits can vary by account, channel, and provider status. Live /pricing remains the source of truth.

API-client smoke-test FAQ

Can I smoke test the gateway before writing SDK code?

Yes. Use Postman or Bruno to verify the base URL, API key, model discovery, selected model ID, and one non-streaming chat request. Then move the same base URL and model ID into SDK or framework code.

Why run /models before /chat/completions?

/models proves the route and account can discover available IDs without prompt content. It also prevents testing a stale or unavailable model name.

Where are the collection files?

The Postman and Bruno collection guide is in the TKEN examples repository. Read the API-client collection guide.

What changed in the release?

The v0.22.0 examples release adds API-client smoke-test assets for Postman and Bruno. Open the v0.22.0 release.

Validate the gateway from an API client first

Create a TKEN key, set https://www.tken.shop/v1 as the base URL, import the Postman or Bruno collection, run /models, then send one small /chat/completions request with a returned model ID.

Create account Developer resources