Simulating realistic Document summarization parameters (25,000 in / 600 out with 10% cache reuse). Perplexity Sonar Deep Research delivers a 79% cost reduction over Claude Fable 5.
| Traffic Volume Tier | Claude Fable 5 Monthly | Perplexity Sonar Deep Research Monthly | Monthly Savings by picking Perplexity Sonar Deep Research |
|---|---|---|---|
| 1,000 reqs/mo (Dev/Testing) | $257.50 | $54.80 | Save $202.70 / mo |
| 10,000 reqs/mo (Small App) | $2,575.00 | $548.00 | Save $2,027.00 / mo |
| 100,000 reqs/mo (Growth Production) | $25,750.00 | $5,480.00 | Save $20,270.00 / mo |
| 1,000,000 reqs/mo (Scale SaaS) | $257,500.00 | $54,800.00 | Save $202,700.00 / mo |
Perplexity Sonar Deep Research is 79% cheaper for Document summarization workloads. At standard Document summarization parameter ratios (25,000 input tokens, 600 output tokens, 10% cache hit), Perplexity Sonar Deep Research costs $0.0548 per request compared to $0.2575 on Claude Fable 5.
Claude Fable 5 offers a context window of 1,000,000 tokens (max output: 128,000), while Perplexity Sonar Deep Research offers 128,000 tokens (max output: 16,384).
At 100,000 requests per month, using Perplexity Sonar Deep Research saves $20,270.00 every month (or $243,240.00 annually) compared to Claude Fable 5.
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.