How to Setup AI Providers
Configure OpenAI, Anthropic Claude, Google Gemini, and Groq for AI-powered question generation, grading, student AI chat, and content creation
How to Setup AI Providers
Course37 uses AI for intelligent question generation, automated essay grading, student AI chat, content suggestions, and more. You can configure multiple providers and choose which ones students and admins use.
Supported Providers
Course37 supports four AI providers. You can enable one or all of them:
| Provider | Models | Best For |
|---|---|---|
| OpenAI | gpt-4o, gpt-4o-mini, gpt-3.5-turbo | General purpose, question generation, grading |
| Anthropic | claude-3-5-sonnet, claude-3-haiku, claude-3-opus | Long-form content, nuanced grading |
| Google Gemini | gemini-1.5-pro, gemini-1.5-flash | Fast generation, translations, cost-effective |
| Groq | llama-3.1-70b, llama-3.1-8b, mixtral-8x7b | Ultra-fast inference, budget-friendly |
Adding an AI Provider
- Go to Settings → Integrations
- Scroll to the AI/LLM Integrations section
- Click the toggle to enable your chosen provider
- Enter the required credentials:
OpenAI
| Field | Description |
|---|---|
| API Key | Your OpenAI API key (starts with sk-) |
| Organization ID | Optional — your OpenAI org ID |
Get your API key from platform.openai.com/api-keys
Anthropic (Claude)
| Field | Description |
|---|---|
| API Key | Your Anthropic API key (starts with sk-ant-) |
Get your API key from console.anthropic.com
Google Gemini
| Field | Description |
|---|---|
| API Key | Your Google AI API key |
Get your API key from aistudio.google.com
Groq
| Field | Description |
|---|---|
| API Key | Your Groq API key (starts with gsk_) |
Get your API key from console.groq.com/keys
Why Groq? Groq runs open-source models (Llama, Mixtral) on custom hardware at extremely fast speeds. It's the most cost-effective option — ideal for student AI chat where volume is high.
- Click Save
All API keys are encrypted at rest and never exposed in the UI after saving.
Configuring AI for Students vs Admins
Course37 lets you assign different providers and models for students and admin/teachers:
Student AI Settings
| Setting | Description |
|---|---|
| Enable Student AI | Toggle AI chat on/off for students |
| Provider | Which provider students use (OpenAI, Anthropic, Gemini, or Groq) |
| Model | The specific model (e.g., gpt-4o-mini, llama-3.1-8b) |
| Fallback Model | Backup model if the primary hits rate limits |
| Custom Prompt | Organization-defined instructions appended to every student AI interaction |
Tip: Use a cheaper, faster model for students (like
gpt-4o-miniorllama-3.1-8bvia Groq) and a more powerful model for admin tasks.
Admin AI Settings
| Setting | Description |
|---|---|
| Enable Admin AI | Toggle AI assistant on/off for admins (enabled by default) |
| Provider | Which provider admins use |
| Model | The specific model for admin AI |
| Custom Prompt | Custom instructions for the admin AI assistant |
To configure:
- Go to Settings → Integrations
- Scroll to Student AI Chat Settings or Admin AI Chat Settings
- Select the provider and model
- Optionally add a custom prompt
- Click Save
AI Token Management
Course37 tracks AI usage via tokens (the units AI providers use to measure text input/output).
Free Platform Tokens
Every organization gets 200,000 free AI tokens per month from the platform. These are used when you haven't configured your own API keys.
| What | Details |
|---|---|
| Monthly Allocation | 200,000 tokens (resets monthly) |
| Usage Tracking | Visible at Settings → Integrations → AI Usage |
| When Exhausted | AI features pause until next month or you add your own API key |
Using Your Own API Keys
When you add your own API key, AI calls go directly to your provider account:
- No platform token limits — usage is billed by your provider
- You control costs — set spending limits in your provider's dashboard
- Per-message tracking — Course37 tracks input/output tokens for every AI message
Token Usage Per Feature
| Feature | Approx. Tokens | Cost (gpt-4o-mini) |
|---|---|---|
| Generate 5 MCQ questions | ~2,000 | ~$0.001 |
| Grade an essay | ~3,000 | ~$0.002 |
| Student AI chat message | ~1,000 | ~$0.0005 |
| Content suggestion | ~500 | ~$0.0003 |
AI-Powered Features
Once configured, these features become available:
| Feature | Description | Where |
|---|---|---|
| Question Generation | AI generates MCQ, CQ, Short Questions from topics or text | Question Bank |
| Essay Grading | AI evaluates written answers against rubrics | Exam Results |
| Student AI Chat | Students can chat with an AI tutor within courses | Student portal |
| Admin AI Assistant | AI assistant for admins to manage content, get insights | Admin dashboard |
| Content Suggestions | AI suggests course descriptions, titles | Course creation |
| Answer Explanation | AI generates explanations for correct answers | Question editor |
| Translation | AI translates content between languages | Content editor |
Cost Comparison
| Provider / Model | Input Cost (per 1M tokens) | Output Cost (per 1M tokens) | Speed | Best For |
|---|---|---|---|---|
OpenAI gpt-4o-mini | $0.15 | $0.60 | Fast | General tasks |
OpenAI gpt-4o | $2.50 | $10.00 | Medium | Complex grading |
Anthropic claude-3-haiku | $0.25 | $1.25 | Fast | Student chat |
Anthropic claude-3-5-sonnet | $3.00 | $15.00 | Medium | Advanced content |
Gemini gemini-1.5-flash | $0.075 | $0.30 | Very Fast | Translations, bulk tasks |
Groq llama-3.1-8b | $0.05 | $0.08 | Ultra Fast | Student chat, high volume |
Groq llama-3.1-70b | $0.59 | $0.79 | Fast | Quality + speed |
Recommendation: Start with Groq (
llama-3.1-8b) for student AI chat (cheapest, fastest) and OpenAI (gpt-4o-mini) for admin tasks and question generation.
Related Articles
- How to Manage AI Tokens — Token allocation and usage tracking
- How to Setup Student AI Chat — Configure AI chat for students
- How to Add Questions — AI question generation
- How to See Results — AI grading
- How to Setup Other Integrations — All integrations