At 8 requests/user/day (totaling 2,400 queries/mo), running Agentic workflow on DeepSeek V3 (Chat) costs $6.586 / month ($0.6586 / user / mo).
Total expected API invoice for 10 Active Users generating 2,400 queries.
Direct inference cost per MAU to model your SaaS pricing tiers and gross margins.
Savings generated by exploiting 65% cache hits on prompt context.
| Model | Provider | Monthly Spend (10 Active Users) | Cost / User / Mo | Annual Spend | Context Limit |
|---|---|---|---|---|---|
| DeepSeek V3 (Chat) (Current) | deepseek | $6.586 | $0.6586 | $79.027 | 64,000 |
| GPT-5.6 Luna | openai | $12.288 | $1.229 | $147.456 | 1,050,000 |
| GPT-5.4 mini | openai | $46.08 | $4.608 | $552.96 | 256,000 |
| GPT-5.4 nano | openai | $12.468 | $1.247 | $149.616 | 128,000 |
| GPT-4o mini | openai | $11.88 | $1.188 | $142.56 | 128,000 |
| Claude Haiku 4.5 | anthropic | $57.84 | $5.784 | $694.08 | 1,000,000 |
At 8 queries per user/day with DeepSeek V3 (Chat), the estimated cost is $0.6586 per monthly active user (MAU). Total monthly bill for 10 Active Users is $6.586.
Assuming a 65% cache hit rate on repeated system and context tokens, prompt caching saves $7.862 per month ($94.349/year) on DeepSeek V3 (Chat).
With a direct COGS of $0.6586 per user/month on DeepSeek V3 (Chat), charging at least $3.293 per user/month ensures an 80%+ software gross margin.