At 8 requests/user/day (totaling 600,000 queries/mo), running Agentic workflow on Mistral Medium 3.5 costs $21,690.00 / month ($8.676 / user / mo).
Total expected API invoice for 2,500 Active Users generating 600,000 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 (2,500 Active Users) | Cost / User / Mo | Annual Spend | Context Limit |
|---|---|---|---|---|---|
| Mistral Medium 3.5 (Current) | mistral | $21,690.00 | $8.676 | $260,280.00 | 256,000 |
| GPT-5.6 Sol | openai | $76,800.00 | $30.72 | $921,600.00 | 1,050,000 |
| GPT-5.6 Cyber | openai | $192,000.00 | $76.80 | $2,304,000.00 | 1,050,000 |
| GPT-5.5 Standard | openai | $76,800.00 | $30.72 | $921,600.00 | 512,000 |
| GPT-5.5 Pro | openai | $460,800.00 | $184.32 | $5,529,600.00 | 512,000 |
| o3-pro (Frontier Reasoning) | openai | $271,200.00 | $108.48 | $3,254,400.00 | 1,000,000 |
At 8 queries per user/day with Mistral Medium 3.5, the estimated cost is $8.676 per monthly active user (MAU). Total monthly bill for 2,500 Active Users is $21,690.00.
Assuming a 65% cache hit rate on repeated system and context tokens, prompt caching saves $21,060.00 per month ($252,720.00/year) on Mistral Medium 3.5.
With a direct COGS of $8.676 per user/month on Mistral Medium 3.5, charging at least $43.38 per user/month ensures an 80%+ software gross margin.