At 4 requests/user/day (totaling 600,000 queries/mo), running Document summarization on Llama 3.1 8B Instruct costs $778.80 / month ($0.1558 / user / mo).
Total expected API invoice for 5,000 Active Users generating 600,000 queries.
Direct inference cost per MAU to model your SaaS pricing tiers and gross margins.
Savings generated by exploiting 10% cache hits on prompt context.
| Model | Provider | Monthly Spend (5,000 Active Users) | Cost / User / Mo | Annual Spend | Context Limit |
|---|---|---|---|---|---|
| Llama 3.1 8B Instruct (Current) | meta | $778.80 | $0.1558 | $9,345.60 | 128,000 |
| GPT-5.6 Luna | openai | $3,162.00 | $0.6324 | $37,944.00 | 1,050,000 |
| GPT-5.4 mini | openai | $11,857.50 | $2.372 | $142,290.00 | 256,000 |
| GPT-5.4 nano | openai | $3,180.00 | $0.636 | $38,160.00 | 128,000 |
| GPT-4o mini | openai | $2,353.50 | $0.4707 | $28,242.00 | 128,000 |
| Claude Haiku 4.5 | anthropic | $15,450.00 | $3.09 | $185,400.00 | 1,000,000 |
At 4 queries per user/day with Llama 3.1 8B Instruct, the estimated cost is $0.1558 per monthly active user (MAU). Total monthly bill for 5,000 Active Users is $778.80.
Assuming a 10% cache hit rate on repeated system and context tokens, prompt caching saves $0.00 per month ($0.00/year) on Llama 3.1 8B Instruct.
With a direct COGS of $0.1558 per user/month on Llama 3.1 8B Instruct, charging at least $0.7788 per user/month ensures an 80%+ software gross margin.