Gemini 3.1 Pro · Google

What does Gemini 3.1 Pro cost to run in a coding agent?

Gemini 3.1 Pro costs about $0.095 per average coding-agent request on our measured traffic, at list prices of $2 per million input and $12 per million output tokens. It holds long contexts well; Artificial Analysis scores Gemini 3.1 Pro Preview 30 on its Intelligence Index. Source: Artificial Analysis (artificialanalysis.ai)

List prices: Gemini API pricing, checked 2026-10-13. Request costs: Standard Compute gateway data, 79,301 coding-agent requests from 75 users, Sep 13 to Oct 12, 2026. How we compute this

Run Gemini 3.1 Pro in your agent

Or let the router send each step to the cheapest model that can do it, on providers we vet for price, uptime and speed. Plans from $19 a month.

See plans

What Gemini 3.1 Pro costs per coding-agent request

Cost at list price	Gemini 3.1 Pro
Prompt under 100K tokens50% of requests, 87% of input cached	$0.032
Prompt of 100K to 272K39% of requests, 88% of input cached	$0.11
Prompt over 272K tokens11% of requests, 86% of input cached	$0.34
Average request	$0.095
100 requests	$9.48
Median session16 requests; a session ends after 10 idle minutes	$0.93
A month of regular usemedian of 20 users active on 5+ days	$135

Each logged request priced at the vendor’s list rates for its own token counts, including cache reads and any higher rate for long prompts. Average output: 698 tokens per request.

Gemini 3.1 Pro benchmarks, speed and latency

Artificial Analysis	Value	Rank
Intelligence Index10 evaluations, 0 to 100	30	29 of 48
Terminal-Bench 4.0agentic coding in a terminal	4.0%	25 of 39
Output speedmedian	144 tokens/s	8 of 44
Time to first answer tokenincludes thinking time	23 s	12 of 44

Source: Artificial Analysis (artificialanalysis.ai). Measured variant: Gemini 3.1 Pro Preview, speed on Google. Fetched October 13, 2026. Rank is among the models on this site, older versions included. Artificial Analysis runs Terminal-Bench 4.0 in its own setup, so its scores differ from the tbench.ai leaderboard.

About Gemini 3.1 Pro

Gemini 3.1 Pro is Google's Pro model, released as a preview on February 19, 2026 with native multimodal input (text, image, file, audio, video) and a 1M-token context. It remains a preview, so Google can change or retire it. Newer models have passed it: Artificial Analysis scores Gemini 3.1 Pro Preview 30 on its Intelligence Index and 4.0% in its Terminal-Bench 4.0 run, while Gemini 3.8 Flash scores 41 and costs about half as much per request in our measurements ($0.048 vs $0.095). It outputs about 144 tokens per second.

Sources: Artificial Analysis

Gemini 3.1 Pro FAQ

How much does Gemini 3.1 Pro cost per coding-agent request?

$0.095 on average at list price, $9.48 per 100 requests, and $0.93 for a median 16-request session. That is 79,301 real coding-agent requests (Sep 13 to Oct 12, 2026) priced at Google's list rates; the median request sent 99,534 input tokens, 87% of input was read from cache, and requests returned 698 output tokens on average.

Does Gemini 3.1 Pro charge more for long prompts?

Yes. Prompts over 200K tokens are billed at $4 per million input and $18 per million output tokens, on every token in the request including cache reads. 21% of the coding-agent requests we measured were that large.

How fast is Gemini 3.1 Pro?

Artificial Analysis measures Gemini 3.1 Pro Preview at 144 output tokens per second, with 23 seconds to the first answer token, thinking time included, on Google's API. Source: Artificial Analysis (artificialanalysis.ai).

Is Gemini 3.1 Pro good for AI agents?

Long multimodal inputs and Gemini CLI setups that still pin 3.1 Pro. On our scoring, which is calibrated to public benchmarks, it rates 5.2/10 for agentic ability and 6.4/10 for output quality. 1M multimodal context.

What is Gemini 3.1 Pro's context window?

Gemini 3.1 Pro has a 1.0M-token context window, large enough to hold sizeable codebases or document sets in a single run. Google is the maker.

Can I self-host Gemini 3.1 Pro?

No. Gemini 3.1 Pro is a proprietary, hosted-only model from Google. You access it through an API, such as OpenRouter, rather than running the weights yourself.

What are the downsides of Gemini 3.1 Pro?

The main trade-offs: Gemini 3.8 Flash scores higher and costs about half as much per request; Prompts over 200K tokens bill at $4/$18 per million; Still a preview model. It's strongest for long multimodal inputs and Gemini CLI setups that still pin 3.1 Pro.