Gemini 3.5 Flash · Google

What does Gemini 3.5 Flash cost to run in a coding agent?

Gemini 3.5 Flash costs about $0.049 per average coding-agent request on our measured traffic, at OpenRouter's listed prices of $1.50 per million input and $9 per million output tokens. It scores 33 on the Artificial Analysis Intelligence Index, 26th of the 48 models we track. Source: Artificial Analysis (artificialanalysis.ai)

List prices: OpenRouter listed price, 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

A newer version is out. Gemini 3.8 Flash →

Run Gemini 3.8 Flash 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 3.5 Flash costs per coding-agent request

Cost at list price	Gemini 3.5 Flash
Prompt under 100K tokens50% of requests, 87% of input cached	$0.024
Prompt of 100K to 272K39% of requests, 88% of input cached	$0.060
Prompt over 272K tokens11% of requests, 86% of input cached	$0.13
Average request	$0.049
100 requests	$4.93
Median session16 requests; a session ends after 10 idle minutes	$0.67
A month of regular usemedian of 20 users active on 5+ days	$75.17

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

Gemini 3.5 Flash benchmarks, speed and latency

Artificial Analysis	Value	Rank
Intelligence Index10 evaluations, 0 to 100	33	26 of 48
Terminal-Bench 4.0agentic coding in a terminal	6.6%	24 of 39
Output speedmedian	197 tokens/s	4 of 44
Time to first answer tokenincludes thinking time	17 s	7 of 44

Source: Artificial Analysis (artificialanalysis.ai). Measured variant: Gemini 3.5 Flash (High), 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.

What to watch for

Gemini 3.8 Flash has replaced it and costs $0.048 per request on our traffic, 2% less. The price is OpenRouter's listed rate on October 13, 2026, which can be a third-party host's; Google's own API may charge differently. It is weak on long agentic coding tasks (see Terminal-Bench 4.0 above), so a cheap request can turn into more steps and retries.

About Gemini 3.5 Flash

Gemini 3.5 Flash was Google's Flash model from May 19, 2026, with a 1M-token context and multimodal input, until Gemini 3.8 Flash replaced it in September. It is fast, at about 197 output tokens per second per Artificial Analysis, but scores 33 on the Artificial Analysis Intelligence Index and 6.6% in its Terminal-Bench 4.0 run, against 41 and 19.7% for Gemini 3.8 Flash. At OpenRouter's listed price ($1.50/$9 per million) it costs about the same per request as Gemini 3.8 Flash at Google's 2027 standard price, and about twice as much as 3.8 Flash at today's introductory price.

Sources: Artificial Analysis

Gemini 3.5 Flash FAQ

How much does Gemini 3.5 Flash cost per coding-agent request?

$0.049 on average at OpenRouter's listed price, $4.93 per 100 requests, and $0.67 for a median 16-request session. That is 79,301 real coding-agent requests (Sep 13 to Oct 12, 2026) priced at OpenRouter's listed 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 3.5 Flash charge more for long prompts?

No. Gemini 3.5 Flash bills the same rate per token up to its 1M-token context window.

How fast is Gemini 3.5 Flash?

Artificial Analysis measures Gemini 3.5 Flash (High) at 197 output tokens per second, with 17 seconds to the first answer token, thinking time included, on Google's API. Source: Artificial Analysis (artificialanalysis.ai).

Is Gemini 3.5 Flash good for AI agents?

Existing Gemini 3.5 Flash setups until they move to 3.8 Flash. On our scoring, which is calibrated to public benchmarks, it rates 5.5/10 for agentic ability and 6.8/10 for output quality. 1M context and multimodal input.

What is Gemini 3.5 Flash's context window?

Gemini 3.5 Flash 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 3.5 Flash?

No. Gemini 3.5 Flash 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 3.5 Flash?

The main trade-offs: Superseded by Gemini 3.8 Flash, which scores higher at a similar or lower price; Weak on long agentic tasks (6.6% in Artificial Analysis's Terminal-Bench 4.0 run). It's strongest for existing Gemini 3.5 Flash setups until they move to 3.8 Flash.