Models / Gemini 3 Pro Preview

Google

Gemini 3 Pro Preview

Gemini 3 Pro Preview is best for writing, and reasoning and knowledge; it is not on sale through OpenRouter, so it is here for reference.

Superseded. Gemini 3.1 Pro Preview is Google's newer model in this line, at $4.50 blended: see Gemini 3.1 Pro Preview or compare the two.

Results 16 Nov 2026 · 63 results · 9 sources

Price per million tokens
Not available through OpenRouter
Speed
Not measured
Intelligence Index
28.0
Artificial Analysis, at high reasoning; 22.3–28.0 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
18 Nov 2025

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 59th of 186Claude Opus 5.5 – – –
Models from developers based in the US 39th of 93Claude Opus 5.5 – – –

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. 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 82nd of 160 available
BenchmarkGemini 3 Pro PreviewRank among availableBest available
tau2-benchtau2 v1.0.1, retail: task success (pass^1) · reported 75.9% 5th of 7 Qwen3.5-397B-A17B 84.4%
tau2-benchtau2 v1.0.1, airline: task success (pass^1) · reported 80.5% 6th of 7 Claude Opus 4.5 84.0%
tau2-benchtau2 v1.0.1, telecom: task success (pass^1) · reported 91.0% 4th of 7 Qwen3.5-397B-A17B 97.8%
Berkeley Function Calling Leaderboard (BFCL) V4BFCL: overall accuracy · reported 72.5% 3rd of 37 Claude Opus 4.5 77.5%
OpenHands IndexOpenHands Index: average · reported 49.0% 19th of 32 Claude Fable 5 81.0%
Vending-Bench 2Vending-Bench 2: money after a year · reported $5,478 21st of 54 GPT-6 Astra $15,515
  • Microsoft STATE-Bench: has not measured this model.
  • Artificial Analysis: has not measured this model.
Professional workReal tasks from banking, consulting and law, business documents and freelance projects 45th of 52 available
BenchmarkGemini 3 Pro PreviewRank among availableBest available
Remote Labor IndexRemote Labor Index: projects done to client standard · reported 1.3% 12th of 13 GPT-6 Astra 20.8%
  • APEX-Agents: has not measured this model.
  • GDP.pdf: has not measured this model.
Reasoning and knowledgeHard questions across science, maths and general knowledge 51st of 188 available
BenchmarkGemini 3 Pro PreviewRank among availableBest available
Artificial AnalysisArtificial Analysis: Artificial Analysis Intelligence Index · reported 28.0 59th of 186 Claude Opus 5.5 57.6
Artificial AnalysisArtificial Analysis: Humanity's Last Exam · reported 39.7% 46th of 187 Claude Opus 5.5 61.4%
Artificial AnalysisArtificial Analysis: GPQA Diamond · reported 90.8% 33rd of 169 GPT-6 Astra 96.3%
WritingWhat people prefer in blind comparisons, and judged writing quality 21st of 170 available
BenchmarkGemini 3 Pro PreviewRank among availableBest available
Arena (formerly LMArena)Arena Text: overall · reported 1,485 15th of 154 Gemini 4 Argon 1,525
UGI LeaderboardUGI: writing score · reported 66.7 21st of 125 Gemini 3.8 Flash 78.6
Sticking to the factsSummarising without inventing things, and factual answers 62nd of 140 available
BenchmarkGemini 3 Pro PreviewRank among availableBest available
Vectara Hallucination LeaderboardVectara: hallucination rate · reported 13.6% 62nd of 72 GPT-5.4 nano 3.1%
Arena (formerly LMArena)Arena Text factuality: overall · reported 1,481 11th of 110 Gemini 4 Argon 1,511
  • SimpleQA Verified (Epoch AI): has not measured this model.
Speed under pressureGood decisions against a real clock (fast chess) Not measured
  • BulletBench: has not measured this model.

Nearest alternatives

Against Gemini 3 Pro Preview, Intelligence Index 28.0. "Similar intelligence" means within 4 points or better.

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 Pro PreviewClaude Opus 5.5
most intelligent available today
Claude Fable 5.2
leads overall
Gemini 4 Argon
Google's best other model
GPT-6.1 Sol
near the top overall
Price per million tokens – $8.00$16.00$4.00$4.00
Tokens a second – 96––53
Intelligence Index 28.0 57.6–52.651.8
Spring Prompt overall – 82907280
Rank among available models, by use case
Product listings – 13th5th11th2nd
Decks from an analysis – 7th4th23rd5th
User surveys – 4th1st15th3rd
Marketing planning – 9th1st13th–
Agents and tool use 82nd of 160 3rd–5th6th
Professional work 45th of 52 3rd––9th
Reasoning and knowledge 51st of 188 1st–4th6th
Writing 21st of 170 5th–1st16th
Sticking to the facts 62nd of 140 3rd–1st20th
Speed under pressure – –8th26th3rd

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

Research and examples

Where it has been measured

Compared with the models available today (it is not on sale itself), in the top quarter (by percentile) for writing, and reasoning and knowledge; in the bottom quarter for professional work.

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

Every result

63 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) · 9 results
MeasureValueRankConfigurationDated
Arena Document: overallArena rating, higher is better 1,451
1,442–1,460
31 of 44 Gemini 3 Pro Preview 13 Sep 2026
Arena Search: overallArena rating, higher is better 1,201
1,196–1,207
9 of 34 Gemini 3 Pro Preview (arena search) 24 Aug 2026
Arena Text factuality: overallArena rating, higher is better 1,481
1,478–1,485
13 of 182 Gemini 3 Pro Preview 2 Oct 2026
Arena Text: business, management and financeArena rating, higher is better 1,474
1,467–1,481
34 of 406 Gemini 3 Pro Preview 2 Oct 2026
Arena Text: creative writingArena rating, higher is better 1,484
1,476–1,493
8 of 411 Gemini 3 Pro Preview 2 Oct 2026
Arena Text: expert promptsArena rating, higher is better 1,498
1,487–1,509
45 of 363 Gemini 3 Pro Preview 2 Oct 2026
Arena Text: instruction followingArena rating, higher is better 1,473
1,466–1,479
32 of 413 Gemini 3 Pro Preview 2 Oct 2026
Arena Text: overallArena rating, higher is better 1,485
1,482–1,489
19 of 413 Gemini 3 Pro Preview 2 Oct 2026
Arena Text: writing, literature and languageArena rating, higher is better 1,481
1,474–1,488
12 of 412 Gemini 3 Pro Preview 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 · 14 results
MeasureValueRankConfigurationDated
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 28.0 130 of 413 Gemini 3 Pro Preview (high reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 22.3 186 of 413 Gemini 3 Pro Preview (low reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 90.8% 56 of 360 Gemini 3 Pro Preview (high reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 88.7% 88 of 360 Gemini 3 Pro Preview (low reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 39.7% 90 of 419 Gemini 3 Pro Preview (high reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 29.5% 152 of 419 Gemini 3 Pro Preview (low reasoning) 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 70.4% 62 of 288 Gemini 3 Pro Preview (high reasoning) 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 49.7% 143 of 288 Gemini 3 Pro Preview (low reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 76.0% 144 of 407 Gemini 3 Pro Preview (high reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 74.0% 163 of 407 Gemini 3 Pro Preview (low reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 41.7% 48 of 282 Gemini 3 Pro Preview (high reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 34.1% 84 of 282 Gemini 3 Pro Preview (low reasoning) 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 87.1% 65 of 286 Gemini 3 Pro Preview (high reasoning) 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 68.1% 133 of 286 Gemini 3 Pro Preview (low reasoning) 20 Oct 2026

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

Berkeley Function Calling Leaderboard (BFCL) V4 · reported by Berkeley Function Calling Leaderboard (BFCL) V4 · 16 results
MeasureValueRankConfigurationDated
BFCL: irrelevance detection% correct, higher is better 77.9% 66 of 109 Gemini 3 Pro Preview (bfcl fc) 16 Dec 2025
BFCL: irrelevance detection% correct, higher is better 85.6% 31 of 109 Gemini 3 Pro Preview (bfcl prompt) 16 Dec 2025
BFCL: memory% correct, higher is better 54.8% 6 of 109 Gemini 3 Pro Preview (bfcl fc) 16 Dec 2025
BFCL: memory% correct, higher is better 61.7% 3 of 109 Gemini 3 Pro Preview (bfcl prompt) 16 Dec 2025
BFCL: multi-turn tasks% correct, higher is better 63.1% 6 of 109 Gemini 3 Pro Preview (bfcl fc) 16 Dec 2025
BFCL: multi-turn tasks% correct, higher is better 60.8% 10 of 109 Gemini 3 Pro Preview (bfcl prompt) 16 Dec 2025
BFCL: overall accuracy% correct, higher is better 68.1% 7 of 109 Gemini 3 Pro Preview (bfcl fc) 16 Dec 2025
BFCL: overall accuracy% correct, higher is better 72.5% 3 of 109 Gemini 3 Pro Preview (bfcl prompt) 16 Dec 2025
BFCL: relevance detection% correct, higher is better 75.0% 62 of 109 Gemini 3 Pro Preview (bfcl fc) 16 Dec 2025
BFCL: relevance detection% correct, higher is better 68.8% 80 of 109 Gemini 3 Pro Preview (bfcl prompt) 16 Dec 2025
BFCL: single-turn calls (curated)% correct, higher is better 85.8% 38 of 109 Gemini 3 Pro Preview (bfcl fc) 16 Dec 2025
BFCL: single-turn calls (curated)% correct, higher is better 90.7% 1 of 109 Gemini 3 Pro Preview (bfcl prompt) 16 Dec 2025
BFCL: single-turn calls (user-contributed)% correct, higher is better 81.7% 6 of 109 Gemini 3 Pro Preview (bfcl fc) 16 Dec 2025
BFCL: single-turn calls (user-contributed)% correct, higher is better 83.1% 2 of 109 Gemini 3 Pro Preview (bfcl prompt) 16 Dec 2025
BFCL: web search% correct, higher is better 68.5% 17 of 109 Gemini 3 Pro Preview (bfcl fc) 16 Dec 2025
BFCL: web search% correct, higher is better 80.0% 7 of 109 Gemini 3 Pro Preview (bfcl prompt) 16 Dec 2025

Not shown: Not general refusal or safety behaviour.

OpenHands Index · reported by OpenHands Index · 6 results
MeasureValueRankConfigurationDated
OpenHands Index: average% resolved, higher is better 49.0% 19 of 34 Gemini 3 Pro Preview (openhands) 30 Jun 2026
OpenHands Index: front end (SWE-Bench Multimodal)% resolved, higher is better 36.8% 11 of 34 Gemini 3 Pro Preview (openhands) 30 Jun 2026
OpenHands Index: greenfield (Commit0)% resolved, higher is better 25.0% 14 of 34 Gemini 3 Pro Preview (openhands) 30 Jun 2026
OpenHands Index: information gathering (GAIA)% resolved, higher is better 44.2% 25 of 34 Gemini 3 Pro Preview (openhands) 30 Jun 2026
OpenHands Index: issue resolution (SWE-Bench)% resolved, higher is better 70.6% 25 of 34 Gemini 3 Pro Preview (openhands) 30 Jun 2026
OpenHands Index: testing (SWT-Bench)% resolved, higher is better 68.6% 17 of 34 Gemini 3 Pro 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.

Remote Labor Index · reported by Remote Labor Index · 1 result
MeasureValueRankConfigurationDated
Remote Labor Index: projects done to client standard% of projects, higher is better 1.3% 13 of 14 Gemini 3 Pro Preview 1 Oct 2026

Not shown: Not speed or cost; a small set of projects, so a few points either way are noise.

tau2-bench · reported by tau2-bench · 8 results
MeasureValueRankConfigurationDated
tau2 v1.0.1, airline: consistency (pass^4)% of tasks, higher is better 66.0% 6 of 8 Gemini 3 Pro Preview (high reasoning, tau2) 2 Mar 2026
tau2 v1.0.1, airline: task success (pass^1)% of tasks, higher is better 80.5% 6 of 8 Gemini 3 Pro Preview (high reasoning, tau2) 2 Mar 2026
tau2 v1.0.1, banking knowledge, terminal retrieval: consistency (pass^4)% of tasks, higher is better 4.1% 3 of 3 Gemini 3 Pro Preview (high reasoning, tau2) 2 Mar 2026
tau2 v1.0.1, banking knowledge, terminal retrieval: task success (pass^1)% of tasks, higher is better 18.0% 3 of 3 Gemini 3 Pro Preview (high reasoning, tau2) 2 Mar 2026
tau2 v1.0.1, retail: consistency (pass^4)% of tasks, higher is better 47.4% 5 of 8 Gemini 3 Pro Preview (high reasoning, tau2) 2 Mar 2026
tau2 v1.0.1, retail: task success (pass^1)% of tasks, higher is better 75.9% 5 of 8 Gemini 3 Pro Preview (high reasoning, tau2) 2 Mar 2026
tau2 v1.0.1, telecom: consistency (pass^4)% of tasks, higher is better 74.6% 3 of 8 Gemini 3 Pro Preview (high reasoning, tau2) 2 Mar 2026
tau2 v1.0.1, telecom: task success (pass^1)% of tasks, higher is better 91.0% 4 of 8 Gemini 3 Pro 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 · 6 results
MeasureValueRankConfigurationDated
UGI: requested-length error% off the requested word count, lower is better 34.0% 278 of 379 Gemini 3 Pro Preview (high reasoning) 18 Nov 2025
UGI: requested-length error% off the requested word count, lower is better 33.0% 276 of 379 Gemini 3 Pro Preview (low reasoning) 18 Nov 2025
UGI: style adherencescore from 0 to 1, higher is better 0.37 103 of 379 Gemini 3 Pro Preview (high reasoning) 18 Nov 2025
UGI: style adherencescore from 0 to 1, higher is better 0.36 155 of 379 Gemini 3 Pro Preview (low reasoning) 18 Nov 2025
UGI: writing scorescore out of 100, higher is better 66.7 54 of 379 Gemini 3 Pro Preview (high reasoning) 18 Nov 2025
UGI: writing scorescore out of 100, higher is better 65.6 61 of 379 Gemini 3 Pro Preview (low reasoning) 18 Nov 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 99.4% 58 of 108 Gemini 3 Pro Preview 22 Sep 2026
Vectara: hallucination rate% of summaries, lower is better 13.6% 89 of 108 Gemini 3 Pro 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 $5,478 21 of 63 Gemini 3 Pro Preview 1 Oct 2026

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

Test your own prompt.

Benchmarks aren't your task. Run your prompt on the models you could use, with every answer checked and costed.