Auto-Coursera — AI-powered Coursera quiz assistant
API Provider Setup
Configure a free AI provider and start answering Coursera quizzes in minutes.
Provider Comparison
Not sure which provider to pick? Here's a side-by-side overview to help you decide.
| Provider | Free Tier | Speed | Image Support | Notes |
|---|---|---|---|---|
| OpenRouter | Varies by model | Medium | Yes (with vision models) | Widest model selection, recommended starting point |
| Google Gemini | Yes (generous) | Fast | Yes | Good free tier, strong general knowledge |
| Groq | Yes (rate limited) | Very fast | No | Fastest inference, good for quick quizzes |
| Cerebras | Yes (rate limited) | Very fast | No | Fast inference, limited model selection |
| NVIDIA NIM | Yes (limited) | Medium | Yes (some models) | Enterprise-focused, requires separate API key |
API endpoints (for reference)
- OpenRouter:
https://openrouter.ai/api/v1 - Google Gemini:
https://generativelanguage.googleapis.com/v1beta/openai/chat/completions - Groq:
https://api.groq.com/openai/v1/chat/completions - Cerebras:
https://api.cerebras.ai/v1/chat/completions - NVIDIA NIM:
https://integrate.api.nvidia.com/v1
Cost Overview
Every supported provider has a free tier. The extension uses each provider's free tier by default — no credit card required, no surprise charges.
We recommend configuring at least two providers. If one goes down or hits its rate limit, the extension automatically fails over to the next.
Providers
OpenRouter Recommended
Setup
- Create an account at openrouter.ai
- Go to Settings → Keys
- Click Create Key
- Copy the key (starts with
sk-or-) - Paste into Auto-Coursera settings (🎓 → gear icon → OpenRouter field)
Google Gemini
Setup
- Go to Google AI Studio
- Click Get API Key
- Click Create API Key
- Copy the key (starts with
AIza) - Paste into Auto-Coursera settings (🎓 → gear icon → Gemini field)
Groq
Setup
- Go to Groq Console
- Navigate to API Keys
- Click Create API Key
- Copy the key (starts with
gsk_) - Paste into Auto-Coursera settings (🎓 → gear icon → Groq field)
Cerebras
Setup
- Go to Cerebras Cloud
- Navigate to API Keys
- Click Create
- Copy the key (starts with
cbs-) - Paste into Auto-Coursera settings (🎓 → gear icon → Cerebras field)
NVIDIA NIM
NVIDIA NIM is particularly well-suited for image-based questions thanks to its vision models. Best used as a secondary provider alongside OpenRouter or Gemini.
Setup
- Go to NVIDIA Build
- Navigate to the API section
- Click Get API Key
- Copy the key (starts with
nvapi-) - Paste into Auto-Coursera settings (🎓 → gear icon → NVIDIA field)
Extension Settings
After adding your API key, you can fine-tune the extension's behavior in the settings panel (click 🎓 → Settings):
Confidence Threshold
A decimal value from 0.0–1.0 that controls the minimum confidence needed to highlight an answer. The default is 0.7. Lower values show more answers; higher values are more selective.
Auto-Select
When enabled, high-confidence answers (green) are clicked automatically. Enabled by default — answers above the confidence threshold are automatically selected.
Auto-Start
When enabled, quiz solving begins as soon as questions are detected on the page. No need to click the FAB.
Provider Priority
Set your preferred primary provider. If it fails or hits a rate limit, the extension automatically tries other configured providers.
API Key Storage
All API keys are encrypted at rest with AES-256-GCM (PBKDF2, 100K iterations) before being stored in Chrome's local storage. This protects against casual inspection of stored data. Keys are sent directly from the extension to the AI provider over HTTPS — they never pass through our servers.
Understanding Confidence
Every answer the AI returns includes a confidence score — a number between 0 and 1 representing the model's self-assessed probability that its answer is correct. It is not a guarantee of accuracy; it's the AI's best estimate.
Confidence Tiers
The extension uses three color-coded tiers to give you a quick visual read:
| Tier | Score Range | Color | Meaning |
|---|---|---|---|
| High | ≥ 0.8 | Green | Model is very confident — likely correct |
| Review | 0.5 – 0.8 | Orange | Plausible but double-check before submitting |
| Low | < 0.5 | Red | Uncertain — treat as a guess |
Auto-Select Threshold
The auto-select threshold (default 0.7)
controls which answers are automatically selected when auto-select is
enabled. Only answers at or above this threshold get clicked;
answers below it are highlighted but left for you to review.
- Raise the threshold (e.g. 0.85) → fewer answers auto-selected, but each one is more likely correct.
- Lower the threshold (e.g. 0.5) → more answers auto-selected, but with higher risk of incorrect picks.
You can adjust it with the confidence slider in settings (🎓 → gear icon → Confidence Threshold).
Rate Limits
To avoid hitting AI provider rate limits, the extension throttles itself using a token-bucket rate limiter.
Extension-Side Limits
- Default: 20 requests per minute. The extension queues requests and releases them at a steady pace.
- On large quizzes (20+ questions), answers may appear in batches as requests are queued and processed — this is normal.
- The rate limit can be adjusted in settings (advanced users only). Setting it too high may trigger provider-side rejections.
Provider-Side Limits
Individual AI providers enforce their own rate limits, which vary by plan and provider. If you see a "rate limit exceeded" error:
- Wait 60 seconds and try again
- Switch to a different provider (the extension does this automatically via failover)
- Consider lowering the extension's rate limit setting if errors persist