At 8 requests/user/day (totaling 24,000 queries/mo), running Agentic workflow on Gemma 2 9B Instruct costs $199.20 / month ($1.992 / user / mo).
Total expected API invoice for 100 Active Users generating 24,000 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 Active Users) | Cost / User / Mo | Annual Spend | Context Limit |
|---|---|---|---|---|---|
| Gemma 2 9B Instruct (Current) | $199.20 | $1.992 | $2,390.40 | 8,192 | |
| GPT-5.6 Luna | openai | $122.88 | $1.229 | $1,474.56 | 1,050,000 |
| GPT-5.4 mini | openai | $460.80 | $4.608 | $5,529.60 | 256,000 |
| GPT-5.4 nano | openai | $124.68 | $1.247 | $1,496.16 | 128,000 |
| GPT-4o mini | openai | $118.80 | $1.188 | $1,425.60 | 128,000 |
| Claude Haiku 4.5 | anthropic | $578.40 | $5.784 | $6,940.80 | 1,000,000 |
At 8 queries per user/day with Gemma 2 9B Instruct, the estimated cost is $1.992 per monthly active user (MAU). Total monthly bill for 100 Active Users is $199.20.
Assuming a 65% cache hit rate on repeated system and context tokens, prompt caching saves $0.00 per month ($0.00/year) on Gemma 2 9B Instruct.
With a direct COGS of $1.992 per user/month on Gemma 2 9B Instruct, charging at least $9.96 per user/month ensures an 80%+ software gross margin.