Getting Started

GlideflowAI is an OpenAI- and Anthropic-compatible gateway that gives one API key access to a curated model catalog.

Choose an integration path

Approach	Best for
OpenAI-compatible base URL	Any language or tool; change one `base_url` value
Anthropic-compatible (Claude Code)	Run GlideflowAI models in Claude Code
MCP server	Let an agent call GlideflowAI models as tools

1. Get an API key

Create an account in the GlideflowAI dashboard, add a balance, and create an API key. Keep it in an environment variable instead of source control.

export GLIDEFLOW_API_KEY="sk-your-key"

2. Set the base URL

For OpenAI-compatible clients, use:

https://api.glideflowai.com/v1

For Anthropic-compatible clients such as Claude Code, use the gateway origin without /v1:

https://api.glideflowai.com

3. Send your first request

Both examples call glm-5.2 through the OpenAI-compatible chat completions endpoint.

cURL

curl https://api.glideflowai.com/v1/chat/completions \
  -H "Authorization: Bearer $GLIDEFLOW_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "glm-5.2",
    "messages": [
      {"role": "user", "content": "Explain token routing in one sentence."}
    ],
    "max_tokens": 512
  }'

Python

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["GLIDEFLOW_API_KEY"],
    base_url="https://api.glideflowai.com/v1",
)

response = client.chat.completions.create(
    model="glm-5.2",
    messages=[
        {"role": "user", "content": "Explain token routing in one sentence."}
    ],
    max_tokens=512,
)

print(response.choices[0].message.content)

Next steps

Authentication — handle keys and request headers
Endpoints — choose the route that matches your client
Tools — configure Claude Code, MCP, SDKs, and other clients
Securing AI Agents — scope keys, tools, spend, and data boundaries