GPT-5.6 Terra is the best value for long documents: it reads a whole book-length file at near-flagship quality and a low price. For the deepest analysis, GPT-6 Astra reads just as much with OpenAI’s strongest reasoning. Gemini 3.7 Flash summarises long text quickly on the Free plan.
Best value
Near-flagship quality at a fraction of the cost, built for long documents and analysis.
For the deepest analysis
OpenAI’s most capable model, for questions that need reasoning across a whole document.
For pulling out details
Follows instructions closely, for tasks like “list every date in this contract”.
Prices are each provider’s list price.
| Model | Plan | Input price | Output price | Reads up to | Speed |
|---|---|---|---|---|---|
| GPT-5.6 Terra | Lite | $2 per 1M tokens | $12 per 1M tokens | 1.1M tokens | Balanced |
| GPT-6 Astra | Pro | $10 per 1M tokens | $50 per 1M tokens | 1.1M tokens | Powerful |
| Gemini 3.7 Flash | Free | $0.75 per 1M tokens | $3.75 per 1M tokens | 1.0M tokens | Balanced |
| GPT-4.1 | Free | $2 per 1M tokens | $8 per 1M tokens | 1M tokens | Balanced |
The deciding number is the context window: how much text a model can read in one go. A model with a window of about a million tokens can take a whole contract set or book at once, and answer questions that depend on page 3 and page 300 together.
The price of reading matters more here than anywhere else, because you pay for every page the model reads. Compare input prices, not just output.
Following instructions precisely matters for extraction. Listing every clause or date is a different skill from summarising, and some models are noticeably better at it.
Yes, on the Lite and Pro plans. On the Free plan you can paste the text in instead.
The models here read about a million tokens at once, roughly 750,000 words. That is several long books, or a full set of contracts.