At 20 requests/user/day (totaling 60,000 queries/mo), running Content generation on QwQ 32B (Reasoner) costs $100.416 / month ($1.004 / user / mo).
Total expected API invoice for 100 Active Users generating 60,000 queries.
Direct inference cost per MAU to model your SaaS pricing tiers and gross margins.
Savings generated by exploiting 30% cache hits on prompt context.
| Model | Provider | Monthly Spend (100 Active Users) | Cost / User / Mo | Annual Spend | Context Limit |
|---|---|---|---|---|---|
| QwQ 32B (Reasoner) (Current) | qwen | $100.416 | $1.004 | $1,204.992 | 128,000 |
| GPT-5.6 Terra | openai | $934.08 | $9.341 | $11,208.96 | 1,050,000 |
| GPT-5.4 Workhorse | openai | $1,167.60 | $11.676 | $14,011.20 | 256,000 |
| o3-mini | openai | $361.68 | $3.617 | $4,340.16 | 200,000 |
| o4-mini | openai | $357.72 | $3.577 | $4,292.64 | 256,000 |
| o1-mini | openai | $361.68 | $3.617 | $4,340.16 | 128,000 |
At 20 queries per user/day with QwQ 32B (Reasoner), the estimated cost is $1.004 per monthly active user (MAU). Total monthly bill for 100 Active Users is $100.416.
Assuming a 30% cache hit rate on repeated system and context tokens, prompt caching saves $5.184 per month ($62.208/year) on QwQ 32B (Reasoner).
With a direct COGS of $1.004 per user/month on QwQ 32B (Reasoner), charging at least $5.021 per user/month ensures an 80%+ software gross margin.