At 200 requests/user/day (totaling 300,000 queries/mo), running Data extraction & tagging on Cerebras — Gemma 4 31B costs $705.75 / month ($14.115 / user / mo).
Total expected API invoice for 50 Active Users generating 300,000 queries.
Direct inference cost per MAU to model your SaaS pricing tiers and gross margins.
Savings generated by exploiting 20% cache hits on prompt context.
| Model | Provider | Monthly Spend (50 Active Users) | Cost / User / Mo | Annual Spend | Context Limit |
|---|---|---|---|---|---|
| Cerebras — Gemma 4 31B (Current) | cerebras | $705.75 | $14.115 | $8,469.00 | 128,000 |
| GPT-5.6 Terra | openai | $1,884.00 | $37.68 | $22,608.00 | 1,050,000 |
| GPT-5.4 Workhorse | openai | $2,355.00 | $47.10 | $28,260.00 | 256,000 |
| o3-mini | openai | $924.00 | $18.48 | $11,088.00 | 200,000 |
| o4-mini | openai | $891.00 | $17.82 | $10,692.00 | 256,000 |
| o1-mini | openai | $924.00 | $18.48 | $11,088.00 | 128,000 |
At 200 queries per user/day with Cerebras — Gemma 4 31B, the estimated cost is $14.115 per monthly active user (MAU). Total monthly bill for 50 Active Users is $705.75.
Assuming a 20% cache hit rate on repeated system and context tokens, prompt caching saves $0.00 per month ($0.00/year) on Cerebras — Gemma 4 31B.
With a direct COGS of $14.115 per user/month on Cerebras — Gemma 4 31B, charging at least $70.575 per user/month ensures an 80%+ software gross margin.