At 8 requests/user/day (totaling 24,000,000 queries/mo), running Agentic workflow on Perplexity Sonar Pro costs $3,420,000.00 / month ($34.20 / user / mo).
Total expected API invoice for 100,000 Active Users generating 24,000,000 queries.
Direct inference cost per MAU to model your SaaS pricing tiers and gross margins.
Savings generated by exploiting 65% cache hits on prompt context.
| Model | Provider | Monthly Spend (100,000 Active Users) | Cost / User / Mo | Annual Spend | Context Limit |
|---|---|---|---|---|---|
| Perplexity Sonar Pro (Current) | perplexity | $3,420,000.00 | $34.20 | $41,040,000.00 | 200,000 |
| GPT-5.6 Terra | openai | $1,228,800.00 | $12.288 | $14,745,600.00 | 1,050,000 |
| GPT-5.4 Workhorse | openai | $1,536,000.00 | $15.36 | $18,432,000.00 | 256,000 |
| o3-mini | openai | $871,200.00 | $8.712 | $10,454,400.00 | 200,000 |
| o4-mini | openai | $699,600.00 | $6.996 | $8,395,200.00 | 256,000 |
| o1-mini | openai | $871,200.00 | $8.712 | $10,454,400.00 | 128,000 |
At 8 queries per user/day with Perplexity Sonar Pro, the estimated cost is $34.20 per monthly active user (MAU). Total monthly bill for 100,000 Active Users is $3,420,000.00.
Assuming a 65% cache hit rate on repeated system and context tokens, prompt caching saves $0.00 per month ($0.00/year) on Perplexity Sonar Pro.
With a direct COGS of $34.20 per user/month on Perplexity Sonar Pro, charging at least $171.00 per user/month ensures an 80%+ software gross margin.