Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.llmgrid.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
Step 1: Create an account

Navigate to https://app.llmgrid.ai Sign up with Google
  1. Select Continue with Google.
  2. Complete the sign-in prompts.
  3. You will be redirected to your workspace.
Sign up with GitHub
  1. Select Continue with GitHub.
  2. Authorize access when prompted.
  3. You will be redirected to your workspace.
Sign up with email address Select Sign up if you want to signup with your email.

​
On-boarding

New registration will take you to on-boarding wizard.

​
Step 2: Add Model

Use this quick start to add your first model so users and applications can begin making requests.
  1. Navigate to Models
In the sidebar, go to AI & Models → Models.
  1. Add New Model
Select Add Model and choose the appropriate provider.
  1. Configure Model Details
Enter the provider model name, assign credentials, and define the public model name exposed to users.
  1. Assign Access Controls
Optionally associate the model with teams, access groups, or budgets to control who can use it.
  1. Test & Save
Use Test Connect to verify connectivity, then save the model. The model becomes immediately available for routing and API calls

​
Step 3: Create a Virtual Key

Virtual Keys are used to authenticate requests and apply governance controls.
  1. Open the Virtual Keys section in the LLMGrid UI.
  2. Select Create New Key.
  3. Provide a name and (optionally) assign:
    • Models
    • Budgets
    • Rate limits
  4. Save the key and copy it securely.
This key will be used as your API token when calling LLMGrid.

​
Step 4: Point Your Application to LLMGrid

Replace the OpenAI base URL with your LLMGrid proxy endpoint in your existing SDK configuration.

​
Python Example

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_LLMGRID_API_KEY",
    base_url="https://api.llmgrid.ai/v1"
)

response = client.chat.completions.create(
    model="gpt-4.1",
    messages=[
        {"role": "user", "content": "Hello from LLMGrid"}
    ]
)

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