At 4 requests/user/day (totaling 300,000 queries/mo), running Document summarization on NVIDIA Nemotron-4 340B costs $11,790.00 / month ($4.716 / user / mo).
Total expected API invoice for 2,500 Active Users generating 300,000 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 (2,500 Active Users) | Cost / User / Mo | Annual Spend | Context Limit |
|---|---|---|---|---|---|
| NVIDIA Nemotron-4 340B (Current) | nvidia | $11,790.00 | $4.716 | $141,480.00 | 128,000 |
| GPT-5.6 Sol | openai | $39,525.00 | $15.81 | $474,300.00 | 1,050,000 |
| GPT-5.6 Cyber | openai | $98,812.50 | $39.525 | $1,185,750.00 | 1,050,000 |
| GPT-5.5 Standard | openai | $39,525.00 | $15.81 | $474,300.00 | 512,000 |
| GPT-5.5 Pro | openai | $237,150.00 | $94.86 | $2,845,800.00 | 512,000 |
| o3-pro (Frontier Reasoning) | openai | $150,900.00 | $60.36 | $1,810,800.00 | 1,000,000 |
At 4 queries per user/day with NVIDIA Nemotron-4 340B, the estimated cost is $4.716 per monthly active user (MAU). Total monthly bill for 2,500 Active Users is $11,790.00.
Assuming a 10% cache hit rate on repeated system and context tokens, prompt caching saves $0.00 per month ($0.00/year) on NVIDIA Nemotron-4 340B.
With a direct COGS of $4.716 per user/month on NVIDIA Nemotron-4 340B, charging at least $23.58 per user/month ensures an 80%+ software gross margin.