At 10 requests/user/day (totaling 30,000 queries/mo), running RAG / search-augmented answers on Claude 3.5 Sonnet costs $621.00 / month ($6.21 / user / mo).
Total expected API invoice for 100 Active Users generating 30,000 queries.
Direct inference cost per MAU to model your SaaS pricing tiers and gross margins.
Savings generated by exploiting 50% cache hits on prompt context.
| Model | Provider | Monthly Spend (100 Active Users) | Cost / User / Mo | Annual Spend | Context Limit |
|---|---|---|---|---|---|
| Claude 3.5 Sonnet (Current) | anthropic | $621.00 | $6.21 | $7,452.00 | 200,000 |
| GPT-5.6 Sol | openai | $1,110.00 | $11.10 | $13,320.00 | 1,050,000 |
| GPT-5.6 Cyber | openai | $2,775.00 | $27.75 | $33,300.00 | 1,050,000 |
| GPT-5.5 Standard | openai | $1,110.00 | $11.10 | $13,320.00 | 512,000 |
| GPT-5.5 Pro | openai | $6,660.00 | $66.60 | $79,920.00 | 512,000 |
| o3-pro (Frontier Reasoning) | openai | $3,840.00 | $38.40 | $46,080.00 | 1,000,000 |
At 10 queries per user/day with Claude 3.5 Sonnet, the estimated cost is $6.21 per monthly active user (MAU). Total monthly bill for 100 Active Users is $621.00.
Assuming a 50% cache hit rate on repeated system and context tokens, prompt caching saves $324.00 per month ($3,888.00/year) on Claude 3.5 Sonnet.
With a direct COGS of $6.21 per user/month on Claude 3.5 Sonnet, charging at least $31.05 per user/month ensures an 80%+ software gross margin.