A standard 50 pages document (~25,000 words / 33,325 tokens) costs $0.003333 to ingest and $0.003483 to generate a comprehensive executive summary using Microsoft Phi-4 (14B).
Feed the entire 50 pages document into context to extract fields, entities, or answer queries.
Ingest 50 pages and generate a structured 500-token executive summary with key takeaways.
Translate or reformat the entire 50 pages document into an equivalent length output.
| Model | Provider | Ingestion (50 pages) | Cached Ingestion | Summary Cost | Context Limit |
|---|---|---|---|---|---|
| Microsoft Phi-4 (14B) (Current) | microsoft | $0.003333 | $0.000833 | $0.003483 | 16,384 |
| GPT-5.6 Luna | openai | $0.0064 | $0.00064 | $0.007 | 1,050,000 |
| GPT-5.4 mini | openai | $0.024 | $0.0024 | $0.0263 | 256,000 |
| GPT-5.4 nano | openai | $0.0064 | $0.00064 | $0.007025 | 128,000 |
| GPT-4o mini | openai | $0.0048 | $0.0024 | $0.0051 | 128,000 |
| Claude Haiku 4.5 | anthropic | $0.033 | $0.0033 | $0.0355 | 1,000,000 |
Assuming a standard single-spaced document with ~500 words per page, a 50 pages document contains approximately 25,000 words, which translates to roughly 33,325 tokens using Microsoft Phi-4 (14B)'s tokenizer. Dense PDF tables, legal boilerplate, and OCR scans may increase this by 20–30%.
Microsoft Phi-4 (14B) has a context window of 16,384 tokens. A 50 pages document consumes only 203.40% of its total window, easily fitting in a single prompt without requiring chunking or vector search.
Ingesting the document and outputting a concise 500-token executive summary costs $0.003483. Using 24-hour async batch API queues drops this cost to $0.001741.
If your workflow repeatedly queries or chats with this same 50 pages document, prompt caching reduces subsequent turn input costs by up to 90% (to $0.000833 per turn instead of $0.003333).