Gemini 2.5 Pro context window: 1,048,576 tokens
Gemini 2.5 Pro has a context window of 1,048,576 tokens and a maximum output of 65,536 tokens per response. The window caps the input the model can attend to in one request. The output cap is a separate limit on how long a single response can be. Gemini 2.5 Flash shares the same 1,048,576-token window at lower rates.
Verified 2026-08-10.
| Model | Context window (tokens) | Max output (tokens) |
|---|---|---|
| Gemini 2.5 Pro | 1,048,576 | 65,536 |
| Gemini 2.5 Flash | 1,048,576 | — |
The 200K pricing tier is the trap
Gemini 2.5 Pro does not have one price. It has two, and the split is set by prompt size. Prompts up to 200,000 input tokens are billed at $1.25 per million input tokens and $10.00 per million output tokens. Prompts above 200,000 input tokens are billed at $2.50 in and $15.00 out per million.
| Prompt size | Input / 1M tokens | Output / 1M tokens |
|---|---|---|
| Up to 200K input tokens | $1.25 | $10.00 |
| Above 200K input tokens | $2.50 | $15.00 |
This is a budgeting trap for long-context workloads. The window advertises 1,048,576 tokens, but the price you likely modeled applies only to the first fifth of it. A pipeline that drifts past 200K input tokens per request doubles its input rate and raises its output rate by half without any code change. If you plan around the big window, plan around the higher tier.
Using the full window
Input billing scales with the tokens you send, so the million-token window also sets a high ceiling on cost per request, and on Gemini 2.5 Pro the per-token rate itself rises past 200K. There is a quality consideration too. Long prompts can degrade retrieval of facts placed in the middle of the window, an effect well documented in long-context evaluations. Test at your real prompt length, and keep key material near the start or end of the prompt when accuracy matters.
The closest alternative
The current Claude models offer a 1,000,000-token window with a 128,000-token output cap and no long-context surcharge, which makes them the direct comparison if the 200K tier split is what concerns you.
More Gemini model details are on the Gemini overview, and full rates are on the Gemini pricing page. For every current model side by side, see the context window comparison.