API Authentication

⚠️ BETA: GMTech's API is currently in beta testing. We welcome your feedback!

GMTech uses API key authentication compatible with OpenAI's format.

Overview

• Method: Bearer token authentication
• Format: Keys start with gmtech_ prefix
• Header: Authorization: Bearer gmtech_your_key
• Scope: Team-based with individual tracking

Getting Your API Key

📸 API Keys Interface

The API Keys page showing the beta badge and key management interface

1. Login to app.gmtech.com
2. Navigate to Team Settings → API Keys
3. Generate a new key with a descriptive name
4. Copy immediately (shown only once)
5. Store securely - never commit to version control

Basic Usage

With Environment Variables (Recommended)

# Set environment variable
export GMTECH_API_KEY="gmtech_your_api_key"

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://app.gmtech.com/api/v1",
    api_key=os.getenv("GMTECH_API_KEY")
)

response = client.chat.completions.create(
    model="gpt-4-turbo",
    messages=[{"role": "user", "content": "Hello!"}]
)

Direct Usage

from openai import OpenAI

client = OpenAI(
    base_url="https://app.gmtech.com/api/v1",
    api_key="gmtech_your_api_key"
)

Rate Limits

Error Handling

from openai import AuthenticationError, RateLimitError

try:
    response = client.chat.completions.create(...)
except AuthenticationError:
    print("Invalid API key")
except RateLimitError:
    print("Rate limit exceeded - slow down requests")

Security Best Practices

✅ Do

• Store keys in environment variables
• Use descriptive key names
• Rotate keys regularly
• Monitor usage in dashboard
• Set appropriate rate limits

❌ Don't

• Hardcode keys in source code
• Commit keys to version control
• Share keys in email/chat
• Use keys in client-side JavaScript
• Ignore usage monitoring

Team Management

All API keys within a team:

• Share unified billing
• Use the same credit pool
• Have individual tracking for attribution
• Can have different rate limits

Troubleshooting

"Invalid API key" Error

• Verify key starts with gmtech_
• Check for whitespace (use .strip())
• Confirm key is active in dashboard

Rate Limit Exceeded

import time
from openai import RateLimitError

def retry_with_backoff(client, **kwargs):
    for attempt in range(3):
        try:
            return client.chat.completions.create(**kwargs)
        except RateLimitError:
            if attempt < 2:
                time.sleep(2 ** attempt)  # 1s, 2s
            else:
                raise

Permission Denied

• Check key permissions in dashboard
• Verify model access for your tier
• Ensure key hasn't been revoked

Key Management

View in dashboard:

• Usage stats: Request count, costs
• Last used: Most recent activity
• Status: Active or revoked
• Permissions: Model access and limits

To revoke a key, visit the API Keys page and click Revoke.

Previous: ← Overview | Next: OpenAI Compatible →