o3-deep-research pricing and specs

Input costs $10 per 1M tokens (845 ₽) and output costs $40 per 1M (3,380 ₽). OpenAI bills the prompt and the completion separately, so the price of a call depends on how long both are. Ruble figures use the Bank of Russia rate of 84.51 ₽ per dollar for September 2026.

List prices are quoted per 1M tokens, so 1,000 tokens cost a thousandth of that: divide $10 for input and $40 for output by 1,000. The same rates in rubles are 845 ₽ and 3,380 ₽ per 1M tokens. Both prompt tokens and completion tokens are billed.

The o3-deep-research context window is 200K tokens (200,000). Everything shares that budget: the system prompt, the dialogue history, attached files and the answer the model writes. The figure comes from the OpenAI model card as of September 2026.

The context window is 200K tokens (200,000) and a single response is capped at 100K tokens, so longer output has to be generated in parts. Rate limits are set by OpenAI per account and depend on your plan rather than on the model, so they are not listed here.

Providers quote their tariffs in US dollars, so the ruble amounts here are a conversion at the official Bank of Russia rate of 84.51 ₽ per dollar. For o3-deep-research that gives 845 ₽ per 1M input tokens and 3,380 ₽ per 1M output tokens. When the rate moves, the ruble price moves with it.

o3-deep-research belongs to the o-series family at OpenAI. Models in the ChatGPT line differ in context size, price per 1M tokens and supported capabilities, so compare them by the numbers on their cards rather than by the name. Every ChatGPT model is listed on the provider page.

Yes, o3-deep-research declares a reasoning mode: it works through intermediate steps before it answers. Those steps add output tokens, so a call with a long reasoning chain costs more and takes longer than a plain completion. Capabilities declared for the model: reasoning, tool calling, image input.

The declared capabilities of o3-deep-research are: reasoning, tool calling, image input. It accepts text, images as input, and every attachment consumes tokens from the shared context window of 200K tokens. The list comes from the OpenAI model card and is refreshed with the catalog as of September 2026.