Groq Setup

Groq provides blazing-fast AI inference powered by custom LPU (Language Processing Unit) hardware, delivering up to 10x faster responses than traditional GPUs.

Getting Your API Key

  1. Visit Groq Console
  2. Sign up or sign in with your account
  3. Navigate to “API Keys” section
  4. Click “Create API Key”
  5. Give your key a name (e.g., “Browser Operator”)
  6. Copy the key - it starts with gsk_...

Free Tier: 14,400 requests/day, no credit card required

Configuring Browser Operator

  1. Open Browser Operator
  2. Click the AI Chat button in the toolbar
  3. Click the Settings icon (⚙️) in the AI Chat panel
  4. Select “Groq” from the Provider dropdown
  5. Paste your API key in the API Key field
  6. Click “Fetch Models” to load available models
  7. Select your preferred model (recommended: llama-3.3-70b-versatile)
  8. Click “Save”

Available Models (October 2025)

ModelContextSpeed (TPS)Cost (per 1M tokens)
llama-4-scout128K594$0.11 / $0.34
llama-4-maverick128K562$0.20 / $0.60
llama-3.3-70b-versatile128K394$0.59 / $0.79
llama-3.1-8b-instant128KUltra Fast$0.05 / $0.08
qwen3-32b131K662$0.29 / $0.59
gpt-oss-20b128K1,000$0.075 / $0.30
gpt-oss-120b128K500$0.15 / $0.60
kimi-k2-0905256K200$1.00 / $3.00

Pricing shown as Input / Output per million tokens

Model Recommendations

For Maximum Speed:

  • gpt-oss-20b - Ultra-fast (1,000 TPS)
  • llama-3.1-8b-instant - Lightning fast

For Complex Reasoning:

  • llama-4-maverick - Best for research and analysis
  • gpt-oss-120b - Strong reasoning capabilities

For Long Context:

  • kimi-k2-0905 - 256K context window

Rate Limits

Free Tier

ModelRequests/MinRequests/DayTokens/Min
Llama 3.1 8B3014,4006,000
Llama 3.3 70B301,00012,000
Llama 4301,0006,000
GPT OSS301,0008,000

Developer Tier

Up to 10x higher limits. Examples:

  • Llama 3.1 8B: 1,000 RPM, 500,000 RPD
  • Most Models: 100-400 RPM, 10,000-200,000 RPD

Troubleshooting

“Invalid API Key”

  • Verify complete key starting with gsk_
  • Check key hasn’t been deleted in Groq Console
  • Try generating a new key

“Rate Limit Exceeded”

  • Check usage at Groq Console
  • Wait for rate limit reset (resets every minute)
  • Use smaller/faster models to reduce token usage
  • Upgrade to Developer tier for higher limits

“Model Not Available”

  • Verify model name is correct
  • Try a different model like llama-3.3-70b-versatile
  • Check status.groq.com for outages

Why Choose Groq?

✅ Choose Groq When:

  • Speed is critical (real-time, conversational AI)
  • You prefer open-source models
  • You need a generous free tier
  • Budget is important

❌ Consider Alternatives When:

  • You specifically need GPT-5 or Claude
  • You need the absolute latest proprietary models

Next Steps

Support