groq-prod-checklist

Groq Production Checklist

Overview

Complete pre-launch checklist for deploying Groq-powered applications to production. Covers API key security, model selection, rate limit planning, fallback strategies, and monitoring setup.

Prerequisites

• Staging environment tested with Groq API
• Groq Developer or Enterprise plan (free tier is not suitable for production)
• Production API key created in console.groq.com
• Monitoring and alerting infrastructure ready

Pre-Deployment Checklist

API Key & Auth

• Production API key stored in secret manager (not .env files)
• Key is NOT shared with development or staging environments
• Key rotation procedure documented and tested
• Pre-commit hook blocks gsk_ pattern in code

Model Selection

• Production model chosen and tested (recommend llama-3.3-70b-versatile)
• Fallback model configured (llama-3.1-8b-instant)
• Deprecated model IDs removed (check deprecations)
• max_tokens set to actual expected output size (not context max)

Rate Limit Planning

• Production rate limits known (check console.groq.com/settings/limits)
• Estimated peak RPM < 80% of limit
• Estimated peak TPM < 80% of limit
• Exponential backoff with retry-after header implemented
• Request queue for burst protection (p-queue or similar)

Error Handling

• All Groq error types caught (Groq.APIError, Groq.APIConnectionError)
• 429 errors retried with backoff
• 5xx errors retried with backoff
• 401 errors trigger alert (key may be revoked)
• Network timeouts configured (default 60s may be too long)
• Circuit breaker pattern for sustained failures

Fallback & Degradation

async function completionWithFallback(messages: any[]) {
  try {
    return await groq.chat.completions.create({
      model: "llama-3.3-70b-versatile",
      messages,
      timeout: 15_000,
    });
  } catch (err: any) {
    if (err.status === 429 || err.status >= 500) {
      console.warn("Groq primary failed, trying fallback model");
      try {
        return await groq.chat.completions.create({
          model: "llama-3.1-8b-instant",
          messages,
          timeout: 10_000,
        });
      } catch {
        console.error("Groq fully unavailable, degrading gracefully");
        return { choices: [{ message: { content: "Service temporarily unavailable. Please try again." } }] };
      }
    }
    throw err;
  }
}

Health Check Endpoint

// /api/health or /healthz
export async function GET() {
  const checks: Record<string, any> = { status: "healthy" };
  const start = performance.now();

  try {
    await groq.chat.completions.create({
      model: "llama-3.1-8b-instant",
      messages: [{ role: "user", content: "OK" }],
      max_tokens: 1,
      temperature: 0,
    });
    checks.groq = { status: "connected", latencyMs: Math.round(performance.now() - start) };
  } catch (err: any) {
    checks.status = "degraded";
    checks.groq = { status: "error", error: err.status || err.message };
  }

  return Response.json(checks, { status: checks.status === "healthy" ? 200 : 503 });
}

Monitoring Setup

• Latency histogram (p50, p95, p99)
• Token throughput counter (tokens/sec by model)
• Error rate by status code (429, 5xx)
• Rate limit remaining gauge (from response headers)
• Cost tracking (tokens * price per million)
• Alert: latency p95 > 1s (Groq normally < 200ms)
• Alert: error rate > 5%
• Alert: rate limit remaining < 10%

Spending Controls

• Monthly spending cap set in Groq Console
• Budget alerts at 50%, 80%, 95%
• Auto-pause enabled when cap is reached

Documentation

• Incident runbook created (see groq-incident-runbook)
• Key rotation SOP documented
• On-call knows how to check status.groq.com
• Rollback procedure tested

Go-Live Verification

set -euo pipefail
# Pre-flight checks
echo "1. Groq API status..."
curl -sf https://status.groq.com > /dev/null && echo "OK" || echo "ISSUE"

echo "2. Production key valid..."
curl -sf https://api.groq.com/openai/v1/models \
  -H "Authorization: Bearer $GROQ_API_KEY_PROD" | jq '.data | length'

echo "3. Health endpoint..."
curl -sf https://your-app.com/api/health | jq .

echo "4. Rate limit headroom..."
curl -si https://api.groq.com/openai/v1/chat/completions \
  -H "Authorization: Bearer $GROQ_API_KEY_PROD" \
  -H "Content-Type: application/json" \
  -d '{"model":"llama-3.1-8b-instant","messages":[{"role":"user","content":"ping"}],"max_tokens":1}' \
  2>/dev/null | grep -i "x-ratelimit-remaining"

Error Handling

Resources

• Groq Status Page
• Groq Rate Limits
• Groq Spend Limits
• Groq Models (check deprecations)

Next Steps

For version upgrades, see groq-upgrade-migration.