At 10 requests/user/day (totaling 15,000,000 queries/mo), running RAG / search-augmented answers on Cerebras — GPT OSS 120B costs $47,625.00 / month ($0.9525 / user / mo).
Total expected API invoice for 50,000 Active Users generating 15,000,000 queries.
Direct inference cost per MAU to model your SaaS pricing tiers and gross margins.
Savings generated by exploiting 50% cache hits on prompt context.
| Model | Provider | Monthly Spend (50,000 Active Users) | Cost / User / Mo | Annual Spend | Context Limit |
|---|---|---|---|---|---|
| Cerebras — GPT OSS 120B (Current) | cerebras | $47,625.00 | $0.9525 | $571,500.00 | 128,000 |
| GPT-5.6 Terra | openai | $222,000.00 | $4.44 | $2,664,000.00 | 1,050,000 |
| GPT-5.4 Workhorse | openai | $277,500.00 | $5.55 | $3,330,000.00 | 256,000 |
| o3-mini | openai | $132,000.00 | $2.64 | $1,584,000.00 | 200,000 |
| o4-mini | openai | $115,500.00 | $2.31 | $1,386,000.00 | 256,000 |
| o1-mini | openai | $132,000.00 | $2.64 | $1,584,000.00 | 128,000 |
At 10 queries per user/day with Cerebras — GPT OSS 120B, the estimated cost is $0.9525 per monthly active user (MAU). Total monthly bill for 50,000 Active Users is $47,625.00.
Assuming a 50% cache hit rate on repeated system and context tokens, prompt caching saves $0.00 per month ($0.00/year) on Cerebras — GPT OSS 120B.
With a direct COGS of $0.9525 per user/month on Cerebras — GPT OSS 120B, charging at least $4.763 per user/month ensures an 80%+ software gross margin.