Compatible (BYOK)

Use cURL with GlideflowAI.

A direct command-line way to inspect and test the compatible API. Connect it through GlideflowAI's OpenAI-compatible API with a key you control.

Get a key

What this connection is for

cURL is a command-line HTTP client that makes a gateway request fully visible: URL, headers, JSON body, and raw response are all under your control. It is for developers troubleshooting an integration before they introduce an SDK, framework, or agent-specific setting. A short cURL command can establish whether a key, model ID, and endpoint work independently of another tool's UI.

Choose cURL for smoke tests, CI diagnostics, and reproducing an HTTP issue in a support conversation. It is not a replacement for an SDK in a production application that needs typed responses, retries, streaming helpers, or structured error handling. Keep command history in mind: pasting a live bearer key directly into a shell can leave it in history.

Use a bearer key to call GlideflowAI's authenticated GET /v1/models endpoint first, then send an exact returned model ID to /v1/chat/completions. Put the key in an environment variable when possible rather than copying it into every command. Save a redacted command in team runbooks so another developer can reproduce the check safely. cURL is an independent project; this is a user-managed connection, not an endorsement or partnership.

Why pair it with GlideflowAI

cURL gives you the clearest way to prove a GlideflowAI key and model work before debugging a higher-level client.

The authenticated models response can be copied directly into an agent or SDK configuration, reducing mistakes caused by display names.

Setup

Use a new API key for the tool, select a model from the catalog, and begin with a small task. Do not put a key in a repository, screenshot, or shared configuration file.

1
List the available models
Confirm the key and endpoint first. GET /v1/models requires authentication.
```
curl https://api.glideflowai.com/v1/models \
  -H "Authorization: Bearer sk-your-key"
```

Send a small chat-completions request

Use an exact ID returned by the model list and leave enough output room for a reasoning model.

curl https://api.glideflowai.com/v1/chat/completions \
  -H "Authorization: Bearer sk-your-key" \
  -H "Content-Type: application/json" \
  -d '{"model":"glm-5.2","messages":[{"role":"user","content":"Say OK"}],"max_tokens":512}'

Need the endpoint basics? Read authentication, browse the model catalog, or compare pricing.

Frequently asked questions

Why test /v1/models before chat completions?

The authenticated models request checks both the key and the catalog, so it rules out a bad model ID before you diagnose a generation request.

How do I avoid putting the key in shell history?

Export GLIDEFLOW_API_KEY in a private shell session and use $GLIDEFLOW_API_KEY in the Authorization header instead of pasting the secret inline.

What output budget should I use for a smoke test?

Use a modest explicit max_tokens value such as 512 so a reasoning model has room to return visible text.

Is this an official integration?

No. The connection is configured by you using a compatible API and does not imply sponsorship, endorsement, or a partnership.

Ready to connect cURL?

Create a key, use the setup above, and verify the connection on a small task before you scale it up.