A standard 100 pages document (~50,000 words / 66,650 tokens) costs $0.02 to ingest and $0.0204 to generate a comprehensive executive summary using Codestral 2501.
Feed the entire 100 pages document into context to extract fields, entities, or answer queries.
Ingest 100 pages and generate a structured 500-token executive summary with key takeaways.
Translate or reformat the entire 100 pages document into an equivalent length output.
| Model | Provider | Ingestion (100 pages) | Cached Ingestion | Summary Cost | Context Limit |
|---|---|---|---|---|---|
| Codestral 2501 (Current) | mistral | $0.02 | $0.002 | $0.0204 | 256,000 |
| GPT-5.3 Codex | openai | $0.112 | $0.0112 | $0.119 | 256,000 |
| DeepSeek Coder V2.5 | deepseek | $0.009331 | $0.000933 | $0.009471 | 128,000 |
| Grok Build 0.1 | xai | $0.0667 | $0.0133 | $0.0677 | 256,000 |
| Qwen 2.5 Coder 32B | qwen | $0.0133 | $0.001333 | $0.0136 | 128,000 |
| DeepInfra — Qwen 2.5 Coder 32B | deepinfra | $0.005332 | $0.001333 | $0.005452 | 128,000 |
Assuming a standard single-spaced document with ~500 words per page, a 100 pages document contains approximately 50,000 words, which translates to roughly 66,650 tokens using Codestral 2501's tokenizer. Dense PDF tables, legal boilerplate, and OCR scans may increase this by 20–30%.
Codestral 2501 has a context window of 256,000 tokens. A 100 pages document consumes only 26.04% 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 $0.0204. Using 24-hour async batch API queues drops this cost to $0.0102.
If your workflow repeatedly queries or chats with this same 100 pages document, prompt caching reduces subsequent turn input costs by up to 90% (to $0.002 per turn instead of $0.02).