Simulating realistic Document summarization parameters (25,000 in / 600 out with 10% cache reuse). Amazon Nova Pro delivers a 94% cost reduction over GPT-5.6 Cyber.
| Traffic Volume Tier | GPT-5.6 Cyber Monthly | Amazon Nova Pro Monthly | Monthly Savings by picking Amazon Nova Pro |
|---|---|---|---|
| 1,000 reqs/mo (Dev/Testing) | $329.375 | $20.42 | Save $308.955 / mo |
| 10,000 reqs/mo (Small App) | $3,293.75 | $204.20 | Save $3,089.55 / mo |
| 100,000 reqs/mo (Growth Production) | $32,937.50 | $2,042.00 | Save $30,895.50 / mo |
| 1,000,000 reqs/mo (Scale SaaS) | $329,375.00 | $20,420.00 | Save $308,955.00 / mo |
Amazon Nova Pro is 94% cheaper for Document summarization workloads. At standard Document summarization parameter ratios (25,000 input tokens, 600 output tokens, 10% cache hit), Amazon Nova Pro costs $0.0204 per request compared to $0.3294 on GPT-5.6 Cyber.
GPT-5.6 Cyber offers a context window of 1,050,000 tokens (max output: 128,000), while Amazon Nova Pro offers 300,000 tokens (max output: 5,000).
At 100,000 requests per month, using Amazon Nova Pro saves $30,895.50 every month (or $370,746.00 annually) compared to GPT-5.6 Cyber.
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.