At 15 requests/user/day (totaling 450,000 queries/mo), running Translation on Mistral Medium 3.5 costs $21,481.875 / month ($21.482 / user / mo).
Translation for a 1,000 active users (seed stage saas) audience means 15,000 requests per day on average, each carrying 5,000 input and 5,500 output tokens at a 15% cacheable share — so the effective per-request cost on Mistral Medium 3.5 is $0.0477, of which $455.625/mo comes back from prompt caching. Per active user that's $21.482/mo — the number to hold against your pricing tier and gross margin. Charging at least $107.409/user/mo keeps gross margin at or above 80%.
Quick answer
For 1,000 Active Users, Translation on Mistral Medium 3.5 is modeled at 450,000 requests per month, costing $21,481.875 per month or $21.482 per active user.
Method & trust
This projection combines the selected active-user tier, requests per user per day, input/output tokens per request, and the modeled cache share. It is an operating estimate, not a provider invoice; production traffic and retries should be measured against it.
Total expected API invoice for 1,000 Active Users generating 450,000 queries.
Direct inference cost per MAU to model your SaaS pricing tiers and gross margins.
Savings generated by exploiting 15% cache hits on prompt context.
| Model | Provider | Monthly Spend (1,000 Active Users) | Cost / User / Mo | Annual Spend | Context Limit |
|---|---|---|---|---|---|
| Mistral Medium 3.5 (Current) | mistral | $21,481.875 | $21.482 | $257,782.50 | 256,000 |
| GPT-5.6 Sol | openai | $83,981.25 | $83.981 | $1,007,775.00 | 1,050,000 |
| GPT-5.6 Cyber | openai | $209,953.125 | $209.953 | $2,519,437.50 | 1,050,000 |
| GPT-5.5 Standard | openai | $83,981.25 | $83.981 | $1,007,775.00 | 512,000 |
| GPT-5.5 Pro | openai | $503,887.50 | $503.888 | $6,046,650.00 | 512,000 |
| o3-pro (Frontier Reasoning) | openai | $236,925.00 | $236.925 | $2,843,100.00 | 1,000,000 |
At 15 queries per user/day with Mistral Medium 3.5, the estimated cost is $21.482 per monthly active user (MAU). Total monthly bill for 1,000 Active Users is $21,481.875.
Assuming a 15% cache hit rate on repeated system and context tokens, prompt caching saves $455.625 per month ($5,467.50/year) on Mistral Medium 3.5.
With a direct COGS of $21.482 per user/month on Mistral Medium 3.5, charging at least $107.409 per user/month ensures an 80%+ software gross margin.