Skip to content
VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Models739Deals30d255Policy30d65, 14 pendingSources1,010

Gemini 3 Flash Preview

GoogleReleased Dec 17, 2025Current

Score basis

Gemini 3 Flash Preview is capable in long context, reasoning, instruction following, and multimodal tasks; and behind the leaders in coding, factuality, agentic tasks, and math. Too few results yet to rate safety or multilingual tasks.

Capability profile

Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.

Capable4 capabilities
  1. Long Context−15.1%2 of 3
  2. Reasoning−21.1%5 of 6
  3. Instruction Following−21.7%1 of 3
  4. Multimodal−23.9%2 of 6
Limited4 capabilities
  1. Coding−26.6%6 of 10
  2. Factuality−27.6%4 of 4
  3. Agentic−36.1%4 of 7
  4. Math−44.3%2 of 5
Not rated2 capabilities

Too few results yet to rate Safety or Multilingual.

Price

$0.50input$3.00outputper million tokens

From Artificial Analysis · 3 providers tracked · All prices

Costs more than 61% of 329 priced models · 3:1 input-to-output blend, log scale

Evidence

105results on66benchmarks

  • 41 independently verified
  • 34 aggregator
  • 30 vendor-reported

From 24 sources · latest Oct 7, 2026 · How verification works

API features

Tool callingStructured outputsReasoning

As listed by OpenRouter

Research

36 papers reference Gemini 3 Flash Preview

Gemini 3 Flash Preview benchmark results

105 results on 66 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.

Long Context

CapableFull long context ranking

15.1% behind the leader2 of 3 ranked benchmarks measured

Show 1 more long context resultHide 1 long context result

Reasoning

CapableFull reasoning ranking

21.1% behind the leader5 of 6 ranked benchmarks measured

Show 10 more reasoning resultsHide 10 reasoning results

Instruction Following

CapableFull instruction following ranking

21.7% behind the leader1 of 3 ranked benchmarks measured

Show 1 more instruction following resultHide 1 instruction following result

Multimodal

CapableFull multimodal ranking

23.9% behind the leader2 of 6 ranked benchmarks measured

Show 6 more multimodal resultsHide 6 multimodal results

Coding

LimitedFull coding ranking

26.6% behind the leader6 of 10 ranked benchmarks measured

Show 6 more coding resultsHide 6 coding results

Factuality

LimitedFull factuality ranking

27.6% behind the leader4 of 4 ranked benchmarks measured

Show 2 more factuality resultsHide 2 factuality results

Agentic

LimitedFull agentic ranking

36.1% behind the leader4 of 7 ranked benchmarks measured

Show 4 more agentic resultsHide 4 agentic results

Math

LimitedFull math ranking

44.3% behind the leader2 of 5 ranked benchmarks measured

Show 4 more math resultsHide 4 math results

More results

Benchmarks outside the capability baskets. They are not ranked against a leader.

Show 39 more resultsHide 39 results

Gemini 3 Flash Preview: common questions

Who makes Gemini 3 Flash Preview?

Gemini 3 Flash Preview is made by Google.

When was Gemini 3 Flash Preview released?

Gemini 3 Flash Preview was released on Dec 17, 2025, according to Artificial Analysis.

What is Gemini 3 Flash Preview good at?

Gemini 3 Flash Preview is capable in long context, reasoning, instruction following, and multimodal tasks; and behind the leaders in coding, factuality, agentic tasks, and math. Too few results yet to rate safety or multilingual tasks.

How much does Gemini 3 Flash Preview cost?

Gemini 3 Flash Preview costs $0.50 per million input tokens and $3.00 per million output tokens, according to Artificial Analysis. We track its price at 3 providers. At a mix of three input tokens to one output token, it costs more than 61% of the 329 priced models we track.

How many benchmarks has Gemini 3 Flash Preview been tested on?

We track 105 results for Gemini 3 Flash Preview on 66 benchmarks from 24 sources, 41 of them independently verified. The latest was recorded on Oct 7, 2026.

Which API features does Gemini 3 Flash Preview support?

OpenRouter lists tool calling, structured outputs, and reasoning for Gemini 3 Flash Preview.

About this record

Where Gemini 3 Flash Preview's numbers come from, and every name it appears under.

Tracked since
Apr 27, 2026
Newest source mention
Sep 10, 2026

Where the results come from

Verification: 105 scores · 41 independently verified · 34 aggregator-attributed · 30 vendor-reported. How these tiers are assigned

From 24 sources on 14 sites. Artificial Analysis supplies 34 of them; the 41 independently verified results come from 10 sites. Bars are coloured by trust tier.

  • artificialanalysis.ai34
  • api.llm-stats.com19
  • deepmind.google9
  • arcprize.org8
  • matharena.ai8
  • datasets-server.huggingface.co4
  • epoch.ai4
  • huggingface.co4
  • raw.githubusercontent.com4
  • lmarena.ai3
  • swebench.com3
  • blog.google2
  • labs.scale.com2
  • simple-bench.com1

Also known as

How our sources name Gemini 3 Flash Preview at each reasoning setting.

SettingShort formAPI id
minimalgemini 3 flash preview (minimal)gemini-3-flash (thinking-minimal)
lowgemini 3 flash preview (low)—
mediumgemini 3 flash preview (medium)—
highgemini 3 flash (high reasoning) gemini 3 flash preview (high) gemini 3 flash (high)—
Also listed asgemini 3 flashgemini 3 flash preview (non-reasoning)gemini 3 flash preview (reasoning)gemini-3-flash-preview-20251217gemini-3-flash-preview:batchgemini-3-flash-reasoninggemini~3~flashgemini-3.* flashgemini 3 flash (non-reasoning)