Simulating realistic Document summarization parameters (25,000 in / 600 out with 10% cache reuse). Mistral Large 3 delivers a 8% cost reduction over Together AI — DeepSeek V4.
| Traffic Volume Tier | Mistral Large 3 Monthly | Together AI — DeepSeek V4 Monthly | Monthly Savings by picking Mistral Large 3 |
|---|---|---|---|
| 1,000 reqs/mo (Dev/Testing) | $12.275 | $13.40 | Save $1.125 / mo |
| 10,000 reqs/mo (Small App) | $122.75 | $134.00 | Save $11.25 / mo |
| 100,000 reqs/mo (Growth Production) | $1,227.50 | $1,340.00 | Save $112.50 / mo |
| 1,000,000 reqs/mo (Scale SaaS) | $12,275.00 | $13,400.00 | Save $1,125.00 / mo |
Mistral Large 3 is 8% cheaper for Document summarization workloads. At standard Document summarization parameter ratios (25,000 input tokens, 600 output tokens, 10% cache hit), Mistral Large 3 costs $0.0123 per request compared to $0.0134 on Together AI — DeepSeek V4.
Mistral Large 3 offers a context window of 1,000,000 tokens (max output: 64,000), while Together AI — DeepSeek V4 offers 1,000,000 tokens (max output: 64,000).
At 100,000 requests per month, using Mistral Large 3 saves $112.50 every month (or $1,350.00 annually) compared to Together AI — DeepSeek V4.
Long-context models pay off here — compare price per 1M tokens at your true document size. Summarize once, store the result; don't re-summarize unchanged documents. For batch backfills, nightly jobs can use cache-friendly request ordering.