Getting Started as an Inference Provider

This guide walks you through setting up your organization as an Inference Provider in 15 minutes.

Prerequisites

Before starting:

• A LangMart account (sign up at https://langmart.ai)
• At least one provider API key (OpenAI, Anthropic, Groq, etc.)
• Understanding of your billing model (org-pays vs member-pays)

Quick Start Overview

Step 1: Create Your Organization

Via Web Interface

1. Log in to your LangMart account
2. Navigate to Organizations in the sidebar
3. Click Create Organization
4. Fill in the details:

5. Click Create

Via API

curl -X POST https://api.langmart.ai/api/organizations \
  -H "Authorization: Bearer <your-api-key>" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Acme AI Services",
    "description": "AI model access for Acme teams",
    "billing_email": "[email protected]"
  }'

Step 2: Add Provider Connections

Connections define how your organization accesses AI model providers.

Adding Your First Connection

1. Go to Connections page
2. Click Add Connection
3. Select provider (e.g., OpenAI)
4. Configure:

5. Click Test Connection to verify
6. Click Save

Recommended Connections

For a full-featured provider setup, add connections for:

Via API

curl -X POST https://api.langmart.ai/api/connections \
  -H "Authorization: Bearer <your-api-key>" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Team OpenAI",
    "provider_id": "openai",
    "endpoint_url": "https://api.openai.com/v1",
    "api_key": "sk-your-openai-key",
    "scope": "organization",
    "billing_mode": "org_pays"
  }'

Step 3: Configure Billing

Add Organization Credits

For org-pays connections, you need credits:

1. Go to Organization > Billing
2. Click Add Credits
3. Enter amount (e.g., $100)
4. Complete payment

Set Spending Limits

Prevent runaway costs:

# Set organization monthly limit
curl -X PUT https://api.langmart.ai/api/organizations/<org_id> \
  -H "Authorization: Bearer <your-api-key>" \
  -H "Content-Type: application/json" \
  -d '{
    "spending_limit_monthly": 1000.00
  }'

Configure Default Member Limits

New members automatically get these limits:

Step 4: Invite Members

Enable Join Requests (Recommended)

Allow users to request access:

1. Go to Organization > Settings
2. Enable Allow Join Requests
3. Optionally enable Auto-Approve for trusted domains

Direct Invitations

Invite specific users:

1. Go to Organization > Members
2. Click Invite Member
3. Enter email address
4. Select role:

5. Click Send Invitation

Via API

curl -X POST https://api.langmart.ai/api/organizations/<org_id>/members/invite \
  -H "Authorization: Bearer <your-api-key>" \
  -H "Content-Type: application/json" \
  -d '{
    "email": "[email protected]",
    "role": "member"
  }'

Step 5: Test Member Access

Test as Admin

Verify models are accessible:

# Using your admin API key
curl -X POST https://api.langmart.ai/v1/chat/completions \
  -H "Authorization: Bearer <your-api-key>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-4o",
    "messages": [{"role": "user", "content": "Hello, testing!"}]
  }'

Test as Member

1. Have a member log in
2. Go to Chat interface
3. Select an organization model
4. Send a test message

Verify Usage Tracking

After tests, check:

1. Organization > Usage - See aggregate usage
2. Analytics > Request Logs - View individual requests

Configuration Summary

After completing setup, you should have:

Next Steps

Expand Your Service

• Add more connections: Support additional providers
• Create connection pools: High availability setup
• Configure PII detection: Enable data protection

Manage at Scale

• Bulk invitations: Import member lists
• Usage reports: Export for billing/compliance
• Alerts: Set up spending notifications

Advanced Configuration

• Organization Setup - Advanced settings
• Billing Models - Detailed billing configuration
• Connection Pools - Load balancing and failover
• Usage Analytics - Monitoring and reporting

Troubleshooting

Connection Test Failed

Member Can't Access Models

Credits Not Deducted

Support

Need help? Contact us:

• Email: [email protected]
• Documentation: Full Provider Guide
• Troubleshooting: Common Issues