# Gemini 3.1 Pro Preview > Gemini 3.1 Pro Preview per million tokens, with a long-context tier and cache read. Agent-loop cost at 0%, 50% and 90% hit, next to Gemini 3.8 Flash. - URL: https://bytescope.dev/en/llm-pricing/gemini-3.1-pro-preview/ - Locale: en · Also: [ja](https://bytescope.dev/ja/llm-pricing/gemini-3.1-pro-preview.md) · [zh-Hant](https://bytescope.dev/zh-Hant/llm-pricing/gemini-3.1-pro-preview.md) - Runs: The rate card is read in this tab. Nothing is uploaded and no model API is called. - Requires: Any modern browser (Chrome, Edge, Firefox, Safari), desktop and mobile. - Updated: 2026-09-08 ## What it does ### Gemini 3.1 Pro Preview The Pro preview is the Google card that publishes a long-context tier. Cache reads are on the card. Cache writes are not. The scenario still models a prefix that has to be written when the hit rate is below one, and those tokens are priced as input, with a mark that the card has no write line. Flash is the peer: no long-context tier, and the same missing write lines. The word preview is in the id because that is what the vendor published. This page does not rename it to a stable alias that the card does not list. If you are deciding whether a prompt might cross the threshold, this is the card that has a threshold. Flash is the card that does not. ### How this card is billed The numbers on this page are not typed in. A checked rate card lives in one table, and every figure you see is read from that table or returned by the cost function the rest of the site calls. The scenario is `agent-loop-50`: fifty turns, one cacheable context that comes back every turn, a little new input that is never cached, and output on every turn. The three totals further down set the cache hit rate at zero, one half, and nine tenths. Hit rate is the share of that repeated context that is read from cache instead of written again. It is not a discount applied after the fact, and it is not the batch factor. Batch, when the card has one, is listed on the card and is left out of these totals on purpose, because a batch job is a different purchase from the interactive loop. When a card publishes no price for a cache line the scenario still has to account for, those tokens are billed as fresh input and the result is marked so. That is a property of the card, not a missing feature of the page. A long-context tier, when the card has one, applies to a single request whose prompt is over the threshold. The fifty short turns of this loop stay on the base tier, which is why a model with a long-context tier still shows one set of loop totals. The worked example below names the day the card was confirmed against the vendor page linked next to it. If you need a different loop, the parent page has the other three scenarios and a slider. ### Worked example Rates on this page were checked against the vendor pricing page on 2026-09-22. 0% cache hit: $4.63 50% cache hit: $2.83 90% cache hit: $1.39 Versus Gemini 3.8 Flash: $1.80 more than Gemini 3.8 Flash. A cache line on this scenario has no published price, so those tokens are billed as input. ### Long context A prompt over 200,000 tokens uses the long-context rates. One request with 200,001 input tokens and 1,000 output tokens, and no cache tokens, costs $0.8180. ## Limits Does not call a model API, does not read agent logs, and does not apply the batch factor to the scenario totals. Rates are the checked card, not a live fetch. ## FAQ ### What does the date on the Gemini 3.1 Pro Preview page mean? It is the day this card was checked against the vendor pricing page. It is not the model's release day, and it is not a promise that the vendor has not changed a number since. When the card changes, the date changes with it. ### Which cache lines does Gemini 3.1 Pro Preview publish? This card publishes cache read and does not publish cache writes. A miss is billed as input and marked, because there is no write line to use. It does publish a long-context tier and a batch factor. The id says preview because that is the id on the card, not a nickname added here. ### Why is Gemini 3.8 Flash the comparison? Flash has no long-context tier and the same missing write lines. The loop delta is the base-tier gap. The threshold only matters on a single long request, which the loop is not.