Groq

WatchThatSite integrates with Groq's API, offering extremely fast LLM inference for latency-sensitive applications.

Get Your API Key

1. Go to Groq Console
2. Sign up or log in with your account
3. Navigate to Keys in the left sidebar
4. Click Create API Key
5. Name it (e.g., "WatchThatSite")
6. Click Create
7. Copy the API key immediately

Add to WatchThatSite

1. Go to Settings > AI Providers
2. Click Add Provider Key
3. Select Groq from the provider dropdown
4. Paste your API key
5. Optionally select specific models
6. Click Save

Recommended Models

Latest Groq models with tool calling:

• Mixtral 8x7B — Fast + capable balance
• LLaMA 2 70B — Large, powerful open model
• LLaMA 2 13B — Smaller, faster

All Available Models

Your account has access to all Groq models supporting tool calling:

1. Go to Settings > AI Providers
2. Click Add Provider Key or Edit existing
3. Models dropdown shows all available models
4. Select specific models to restrict, or leave empty for all

Key Features

• Ultra-fast inference: 500+ tokens/second throughput
• Low latency: Sub-100ms response times
• Cost-effective: Competitive pricing for speed
• Open-source models: Based on popular open models
• No rate limits: Generally higher limits than competitors

Pricing

Groq charges per 1M tokens:

• Input tokens: Price per 1M input tokens
• Output tokens: Price per 1M output tokens (typically 2x input cost)

Free tier available with rate limits. Check Groq pricing for current rates.

Use Cases

Groq excels for:

• Real-time applications: High-speed responses needed
• Chat applications: Low-latency conversational AI
• Streaming responses: Fast token generation
• Cost-sensitive projects: Efficient pricing at scale

Troubleshooting

"Invalid API Key"

• Verify complete key is pasted without spaces
• Ensure key hasn't been revoked
• Check in Groq console the key is active

Rate Limits

Free tier has usage limits:

• Upgrade account for higher limits
• Implement request queuing
• Use batch processing for non-urgent tasks

"Model Not Found"

• Verify model is currently available
• Some models are beta or region-restricted
• Check Groq console for available models
• Request access to new models if needed

High Latency

While Groq is fast, latency depends on:

• Model size
• Request complexity
• Network conditions
• Server load

For optimal results:

• Use smaller models (Mixtral) for speed
• Enable streaming for incremental results
• Use connection pooling