At 6 requests/user/day (totaling 1,800 queries/mo), running Customer support chatbot on Microsoft Phi-3.5 MoE costs $1.323 / month ($0.1323 / user / mo).
Total expected API invoice for 10 Active Users generating 1,800 queries.
Direct inference cost per MAU to model your SaaS pricing tiers and gross margins.
Savings generated by exploiting 70% cache hits on prompt context.
| Model | Provider | Monthly Spend (10 Active Users) | Cost / User / Mo | Annual Spend | Context Limit |
|---|---|---|---|---|---|
| Microsoft Phi-3.5 MoE (Current) | microsoft | $1.323 | $0.1323 | $15.876 | 128,000 |
| GPT-5.6 Luna | openai | $1.222 | $0.1222 | $14.666 | 1,050,000 |
| GPT-5.4 mini | openai | $4.583 | $0.4583 | $54.999 | 256,000 |
| GPT-5.4 nano | openai | $1.254 | $0.1254 | $15.044 | 128,000 |
| GPT-4o mini | openai | $0.9923 | $0.0992 | $11.907 | 128,000 |
| Claude Haiku 4.5 | anthropic | $5.481 | $0.5481 | $65.772 | 1,000,000 |
At 6 queries per user/day with Microsoft Phi-3.5 MoE, the estimated cost is $0.1323 per monthly active user (MAU). Total monthly bill for 10 Active Users is $1.323.
Assuming a 70% cache hit rate on repeated system and context tokens, prompt caching saves $0.00 per month ($0.00/year) on Microsoft Phi-3.5 MoE.
With a direct COGS of $0.1323 per user/month on Microsoft Phi-3.5 MoE, charging at least $0.6615 per user/month ensures an 80%+ software gross margin.