Models / Gemini 3 Flash Preview

Google

Gemini 3 Flash Preview

Gemini 3 Flash Preview is best for writing; a strong pick at a modest price. The cheapest model of similar intelligence is DeepSeek-V4-Flash (0423), at $0.15 per million tokens against $1.12.

Superseded. Gemini 3.8 Flash is Google's newer model in this line, at $3.00 blended: see Gemini 3.8 Flash or compare the two.

Results 16 Nov 2026 · 60 results · 9 sources

Test Gemini 3 Flash Preview on your prompt → It runs beside the models you could use instead, with every answer checked and costed.

Price per million tokens
$0.50 in · $3.00 out
Typical host on OpenRouter, 10 Nov 2026 · compare prices
Speed
Not measured
Intelligence Index
26.3
Artificial Analysis, at thinking reasoning; 17.9–26.3 across 2 settings
Spring Prompt overall
Not ranked yet
needs our benchmarks and two groups of results
Developer
Google, based in the US
Weights not published
Released
17 Dec 2025
1,048,576 tokens of context

Where it stands among the models you could choose

Available models grouped by price, intelligence, speed and developer location. Retired models are excluded.

AmongIntelligence IndexSpring Prompt overallPrice, blendedSpeed
Models available today 66th of 185Claude Opus 5.5 – $1.12 blended, 58th percentilecheapest: Mistral Nemo –
Models priced $1–3 per million tokens 16th of 34Muse Spark 1.3 – sets this group –
Models with similar intelligence sets this group – $1.12 blended, 40th percentilecheapest: DeepSeek-V4-Flash (0423) –
Models from developers based in the US 41st of 92Claude Opus 5.5 – $1.12 blended, 41st percentilecheapest: Gemma 3 4B –

The name under each rank is the leader of that group; green is the top quarter of the group and red the bottom quarter, by rank. Price is shown as a percentile: the share of the group that costs less, so lower is cheaper. "Sets this group" marks the measure the group is defined by. Spring Prompt overall is our score out of 100 across every source (how it works). Price is the developer's list price (or the typical OpenRouter host where we haven't read one) for three input tokens to one output; intelligence and speed are from Artificial Analysis (data sourced from Artificial Analysis); "similar intelligence" means within 4 points on its Intelligence Index. Developer locations are where each company is based, not where a model is served. Groups under 3 models are not ranked.

How good is it, and for what?

Each use case ranks the available models its benchmarks measured, then those priced $1–3 per million tokens. The bar shows its percentile in that field, best to the right. Open a row for the results behind it.

Product listingsTurning a sparse product feed and photos into listings that can go live Not measured
  • CatalogBench: has not measured this model.
Decks from an analysisTurning a finished analysis into a deck you could present as it is Not measured
  • DeckBench: has not measured this model.
User surveysPlanning a user survey and reading its results without being misled Not measured
  • SurveyBench: has not measured this model.
Marketing planningPlanning a year of ad spend without overspending Not measured
  • ROASBench: has not measured this model.
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 85th of 159 available · 20th of 31 at its price
BenchmarkGemini 3 Flash PreviewRank among availableBest available
tau2-benchtau2 v1.0.1, retail: task success (pass^1) · reported 76.8% 4th of 6 Qwen3.5-397B-A17B 84.4%
tau2-benchtau2 v1.0.1, airline: task success (pass^1) · reported 82.5% 3rd of 6 Claude Opus 4.5 84.0%
tau2-benchtau2 v1.0.1, telecom: task success (pass^1) · reported 91.2% 3rd of 6 Qwen3.5-397B-A17B 97.8%
OpenHands IndexOpenHands Index: average · reported 49.0% 19th of 31 Claude Fable 5 81.0%
Artificial AnalysisArtificial Analysis: τ-bench banking · reported 20.8% 45th of 104 Grok 4.6 50.7%
Vending-Bench 2Vending-Bench 2: money after a year · reported $3,635 34th of 53 GPT-6 Astra $15,515
  • Berkeley Function Calling Leaderboard (BFCL) V4: pending. Its latest results are dated 16 Dec 2025, before this model was released.
  • Microsoft STATE-Bench: has not measured this model.
Professional workReal tasks from banking, consulting and law, business documents and freelance projects 46th of 51 available · 10th of 12 at its price
BenchmarkGemini 3 Flash PreviewRank among availableBest available
GDP.pdfGDP.pdf: professional document tasks · reported 10.0% 29th of 31 GPT-5.6 Sol 30.7%
  • APEX-Agents: has not measured this model.
  • Remote Labor Index: has not measured this model.
Reasoning and knowledgeHard questions across science, maths and general knowledge 60th of 187 available · 15th of 35 at its price
BenchmarkGemini 3 Flash PreviewRank among availableBest available
Artificial AnalysisArtificial Analysis: Artificial Analysis Intelligence Index · reported 26.3 66th of 185 Claude Opus 5.5 57.6
Artificial AnalysisArtificial Analysis: Humanity's Last Exam · reported 36.6% 56th of 186 Claude Opus 5.5 61.4%
Artificial AnalysisArtificial Analysis: GPQA Diamond · reported 89.8% 41st of 168 GPT-6 Astra 96.3%
WritingWhat people prefer in blind comparisons, and judged writing quality 28th of 169 available · 6th of 28 at its price
BenchmarkGemini 3 Flash PreviewRank among availableBest available
Arena (formerly LMArena)Arena Text: overall · reported 1,473 30th of 153 Gemini 4 Argon 1,525
UGI LeaderboardUGI: writing score · reported 67.7 16th of 124 Gemini 3.8 Flash 78.6
Sticking to the factsSummarising without inventing things, and factual answers 53rd of 139 available · 8th of 26 at its price
BenchmarkGemini 3 Flash PreviewRank among availableBest available
Vectara Hallucination LeaderboardVectara: hallucination rate · reported 13.5% 61st of 71 GPT-5.4 nano 3.1%
Arena (formerly LMArena)Arena Text factuality: overall · reported 1,468 23rd of 109 Gemini 4 Argon 1,511
SimpleQA Verified (Epoch AI)SimpleQA Verified: correct answers · reported 66.8% 10th of 68 GPT-6 Astra 75.6%
Speed under pressureGood decisions against a real clock (fast chess) Not measured
  • BulletBench: has not measured this model.

Nearest alternatives

Against Gemini 3 Flash Preview at $1.12 blended, Intelligence Index 26.3. "Similar intelligence" means within 4 points or better.

Where to run it

2 hosts on OpenRouter at the standard service tier, with a median of $1.12 blended per million tokens. Cheapest first.

HostQuantisationInputOutputBlended
Googlenot stated$0.50$3.00$1.12
Google AI Studionot stated$0.50$3.00$1.12

Other service tiers, not counted above: Google flex $0.25 in, $1.50 out; Google AI Studio flex $0.25 in, $1.50 out; Google priority $0.90 in, $5.40 out; Google AI Studio priority $0.90 in, $5.40 out. Flex and batch are cheaper but slower or queued; priority and fast cost more.

Prices per million tokens from OpenRouter's public endpoint list, checked 10 Nov 2026.

Against the alternatives

The models you would most likely weigh it against: the leaders of the groups above.

Scroll sideways to see every alternative.

Gemini 3 Flash PreviewMuse Spark 1.3
most intelligent at $1–3 per million tokens
DeepSeek-V4-Flash (0423)
cheapest with similar intelligence
Claude Opus 5.5
most intelligent available today
Price per million tokens $1.12 $2.00$0.15$8.00
Tokens a second – 254–96
Intelligence Index 26.3 48.124.457.6
Spring Prompt overall – 59–82
Rank among available models, by use case
Product listings – 17th–13th
Decks from an analysis – 25th–7th
User surveys – ––4th
Marketing planning – 25th–9th
Agents and tool use 85th of 159 13th59th3rd
Professional work 46th of 51 8th–3rd
Reasoning and knowledge 60th of 187 11th69th1st
Writing 28th of 169 7th63rd5th
Sticking to the facts 53rd of 139 11th81st3rd
Speed under pressure – 33rd––

Shaded figures are better than Gemini 3 Flash Preview's. A dash means no result.

Where it has been measured

16th most intelligent of the 34 models priced $1–3 per million tokens. Among models available today, in the top quarter (by percentile) for writing; in the bottom quarter for professional work.

Artificial Analysis lists it but has not published a speed for it yet

Every result

60 published results from 9 sources, each in the source's own units, with the configuration that produced it.

Arena (formerly LMArena) · reported by Arena (formerly LMArena) · 16 results
MeasureValueRankConfigurationDated
Arena Document: overallArena rating, higher is better 1,439
1,430–1,449
36 of 44 Gemini 3 Flash Preview 13 Sep 2026
Arena Search: overallArena rating, higher is better 1,191
1,187–1,196
19 of 34 Gemini 3 Flash Preview (arena search) 24 Aug 2026
Arena Text factuality: overallArena rating, higher is better 1,468
1,464–1,471
32 of 182 Gemini 3 Flash Preview 2 Oct 2026
Arena Text factuality: overallArena rating, higher is better 1,451
1,449–1,454
67 of 182 Gemini 3 Flash Preview (minimal reasoning) 2 Oct 2026
Arena Text: business, management and financeArena rating, higher is better 1,468
1,460–1,477
40 of 406 Gemini 3 Flash Preview 2 Oct 2026
Arena Text: business, management and financeArena rating, higher is better 1,454
1,449–1,460
70 of 406 Gemini 3 Flash Preview (minimal reasoning) 2 Oct 2026
Arena Text: creative writingArena rating, higher is better 1,457
1,448–1,466
29 of 411 Gemini 3 Flash Preview 2 Oct 2026
Arena Text: creative writingArena rating, higher is better 1,447
1,441–1,453
45 of 411 Gemini 3 Flash Preview (minimal reasoning) 2 Oct 2026
Arena Text: expert promptsArena rating, higher is better 1,496
1,484–1,509
50 of 363 Gemini 3 Flash Preview 2 Oct 2026
Arena Text: expert promptsArena rating, higher is better 1,463
1,456–1,470
98 of 363 Gemini 3 Flash Preview (minimal reasoning) 2 Oct 2026
Arena Text: instruction followingArena rating, higher is better 1,457
1,450–1,465
53 of 413 Gemini 3 Flash Preview 2 Oct 2026
Arena Text: instruction followingArena rating, higher is better 1,443
1,438–1,448
79 of 413 Gemini 3 Flash Preview (minimal reasoning) 2 Oct 2026
Arena Text: overallArena rating, higher is better 1,473
1,468–1,477
42 of 413 Gemini 3 Flash Preview 2 Oct 2026
Arena Text: overallArena rating, higher is better 1,458
1,455–1,461
61 of 413 Gemini 3 Flash Preview (minimal reasoning) 2 Oct 2026
Arena Text: writing, literature and languageArena rating, higher is better 1,461
1,453–1,469
35 of 412 Gemini 3 Flash Preview 2 Oct 2026
Arena Text: writing, literature and languageArena rating, higher is better 1,448
1,443–1,453
56 of 412 Gemini 3 Flash Preview (minimal reasoning) 2 Oct 2026

Not shown: Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.

Artificial Analysis · reported by Artificial Analysis · 15 results
MeasureValueRankConfigurationDated
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 17.9 229 of 413 Gemini 3 Flash Preview (no reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 26.3 143 of 413 Gemini 3 Flash Preview (thinking reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 81.2% 166 of 360 Gemini 3 Flash Preview (no reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 89.8% 68 of 360 Gemini 3 Flash Preview (thinking reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 15.0% 232 of 419 Gemini 3 Flash Preview (no reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 36.6% 110 of 419 Gemini 3 Flash Preview (thinking reasoning) 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 55.1% 120 of 288 Gemini 3 Flash Preview (no reasoning) 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 78.0% 11 of 288 Gemini 3 Flash Preview (thinking reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 55.3% 262 of 407 Gemini 3 Flash Preview (no reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 78.0% 117 of 407 Gemini 3 Flash Preview (thinking reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 31.8% 97 of 282 Gemini 3 Flash Preview (no reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 38.6% 58 of 282 Gemini 3 Flash Preview (thinking reasoning) 20 Oct 2026
Artificial Analysis: τ-bench banking% of tasks, higher is better 20.8% 85 of 175 Gemini 3 Flash Preview (thinking reasoning) 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 43.3% 176 of 286 Gemini 3 Flash Preview (no reasoning) 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 80.4% 101 of 286 Gemini 3 Flash Preview (thinking reasoning) 20 Oct 2026

Not shown: Not business work, and a blend: read the parts for any one task.

GDP.pdf · reported by GDP.pdf · 2 results
MeasureValueRankConfigurationDated
GDP.pdf: professional document tasks% of rubric, higher is better 10.0% 43 of 49 Gemini 3 Flash Preview 6 Jun 2026
GDP.pdf: professional document tasks% of rubric, higher is better 10.0% 43 of 49 Gemini 3 Flash Preview (high reasoning) 1 Oct 2026

Not shown: Not your documents; 100 tasks, so small differences are noise.

OpenHands Index · reported by OpenHands Index · 6 results
MeasureValueRankConfigurationDated
OpenHands Index: average% resolved, higher is better 49.0% 20 of 34 Gemini 3 Flash Preview (openhands) 30 Jun 2026
OpenHands Index: front end (SWE-Bench Multimodal)% resolved, higher is better 22.1% 30 of 34 Gemini 3 Flash Preview (openhands) 30 Jun 2026
OpenHands Index: greenfield (Commit0)% resolved, higher is better 18.8% 21 of 34 Gemini 3 Flash Preview (openhands) 30 Jun 2026
OpenHands Index: information gathering (GAIA)% resolved, higher is better 58.8% 19 of 34 Gemini 3 Flash Preview (openhands) 30 Jun 2026
OpenHands Index: issue resolution (SWE-Bench)% resolved, higher is better 74.6% 13 of 34 Gemini 3 Flash Preview (openhands) 30 Jun 2026
OpenHands Index: testing (SWT-Bench)% resolved, higher is better 70.7% 11 of 34 Gemini 3 Flash Preview (openhands) 30 Jun 2026

Not shown: Not the model alone: the score is for the model inside the OpenHands agent, at the SDK version shown.

SimpleQA Verified (Epoch AI) · reported by SimpleQA Verified (Epoch AI) · 1 result
MeasureValueRankConfigurationDated
SimpleQA Verified: correct answers% of questions, higher is better 66.8% 10 of 85 Gemini 3 Flash Preview (high reasoning) 27 Aug 2026

Not shown: Not answers grounded in your documents; tests what the model remembers.

tau2-bench · reported by tau2-bench · 8 results
MeasureValueRankConfigurationDated
tau2 v1.0.1, airline: consistency (pass^4)% of tasks, higher is better 68.0% 4 of 8 Gemini 3 Flash Preview (high reasoning, tau2) 2 Mar 2026
tau2 v1.0.1, airline: task success (pass^1)% of tasks, higher is better 82.5% 3 of 8 Gemini 3 Flash Preview (high reasoning, tau2) 2 Mar 2026
tau2 v1.0.1, banking knowledge, terminal retrieval: consistency (pass^4)% of tasks, higher is better 7.2% 2 of 3 Gemini 3 Flash Preview (high reasoning, tau2) 2 Mar 2026
tau2 v1.0.1, banking knowledge, terminal retrieval: task success (pass^1)% of tasks, higher is better 27.3% 1 of 3 Gemini 3 Flash Preview (high reasoning, tau2) 2 Mar 2026
tau2 v1.0.1, retail: consistency (pass^4)% of tasks, higher is better 51.8% 2 of 8 Gemini 3 Flash Preview (high reasoning, tau2) 2 Mar 2026
tau2 v1.0.1, retail: task success (pass^1)% of tasks, higher is better 76.8% 4 of 8 Gemini 3 Flash Preview (high reasoning, tau2) 2 Mar 2026
tau2 v1.0.1, telecom: consistency (pass^4)% of tasks, higher is better 70.2% 5 of 8 Gemini 3 Flash Preview (high reasoning, tau2) 2 Mar 2026
tau2 v1.0.1, telecom: task success (pass^1)% of tasks, higher is better 91.2% 3 of 8 Gemini 3 Flash Preview (high reasoning, tau2) 2 Mar 2026

Not shown: Not your policies or customers: the tasks and policies are tau2's own, and the customer is simulated by gpt-5.2.

UGI Leaderboard · reported by UGI Leaderboard · 9 results
MeasureValueRankConfigurationDated
UGI: requested-length error% off the requested word count, lower is better 26.0% 241 of 379 Gemini 3 Flash Preview (high reasoning) 23 Dec 2025
UGI: requested-length error% off the requested word count, lower is better 23.0% 216 of 379 Gemini 3 Flash Preview (medium reasoning) 24 Dec 2025
UGI: requested-length error% off the requested word count, lower is better 16.0% 135 of 379 Gemini 3 Flash Preview (minimal reasoning) 22 Dec 2025
UGI: style adherencescore from 0 to 1, higher is better 0.37 112 of 379 Gemini 3 Flash Preview (high reasoning) 23 Dec 2025
UGI: style adherencescore from 0 to 1, higher is better 0.37 120 of 379 Gemini 3 Flash Preview (medium reasoning) 24 Dec 2025
UGI: style adherencescore from 0 to 1, higher is better 0.34 233 of 379 Gemini 3 Flash Preview (minimal reasoning) 22 Dec 2025
UGI: writing scorescore out of 100, higher is better 63.3 83 of 379 Gemini 3 Flash Preview (high reasoning) 23 Dec 2025
UGI: writing scorescore out of 100, higher is better 67.7 44 of 379 Gemini 3 Flash Preview (medium reasoning) 24 Dec 2025
UGI: writing scorescore out of 100, higher is better 65.1 66 of 379 Gemini 3 Flash Preview (minimal reasoning) 22 Dec 2025

Not shown: Not other format limits such as character counts or bullet counts.

Vectara Hallucination Leaderboard · reported by Vectara Hallucination Leaderboard · 2 results
MeasureValueRankConfigurationDated
Vectara: answer rate% of documents, higher is better 99.8% 34 of 108 Gemini 3 Flash Preview 22 Sep 2026
Vectara: hallucination rate% of summaries, lower is better 13.5% 88 of 108 Gemini 3 Flash Preview 22 Sep 2026

Not shown: Not a quality score: a low rate usually means content filters were triggered, and hallucination rates are measured on answered documents only.

Vending-Bench 2 · reported by Vending-Bench 2 · 1 result
MeasureValueRankConfigurationDated
Vending-Bench 2: money after a yearUS dollars, higher is better $3,635 40 of 63 Gemini 3 Flash Preview 1 Oct 2026

Not shown: Not a real business; one simulated market with set rules.

Will Gemini 3 Flash Preview work for your product?

Benchmarks aren't your task. Paste the prompt your product uses and run it beside the models you could use instead, with every answer checked and costed.

New here? Your email starts a free trial. No credit card needed.