Models / Gemini 3.1 Pro Preview

Google

Gemini 3.1 Pro Preview

Gemini 3.1 Pro Preview is best for writing, and reasoning and knowledge; pick it when quality matters more than cost. The cheapest model of similar intelligence is MiMo-V2.6-Flash, at $0.18 per million tokens against $4.50.

Results 16 Nov 2026 · 138 results · 15 sources

Test Gemini 3.1 Pro Preview on your prompt → It runs beside the models you could use instead, with every answer checked and costed.

Price per million tokens
$2.00 in · $12.00 out
Google's list price, 10 Nov 2026 · compare prices$0.20 cached input
Speed
130 tokens a second; on its test prompt the answer starts after 14.7 s, thinking time included
Artificial Analysis, on its usual API
Intelligence Index
29.7
Artificial Analysis
Spring Prompt overall
20th of 36 score 46 of 100 · how it works
Spring Prompt Score
44 of 100 18th of 32 · 5 of 5 benchmarks · our own tests only
Developer
Google, based in the US
Weights not published
Released
19 Feb 2026
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 54th of 185Claude Opus 5.5 20th of 35Claude Fable 5.2 $4.50 blended, 87th percentilecheapest: Mistral Nemo 30th of 65Nemotron 3.5 Super VL
Models priced $3–10 per million tokens 24th of 38Claude Opus 5.5 12th of 15Claude Opus 5.5 sets this group 7th of 15Mistral Medium 3.5
Models with similar intelligence sets this group – $4.50 blended, 75th percentilecheapest: Nemotron 3.5 Super VL 5th of 8Nemotron 3.5 Super VL
Models writing 100–200 tokens a second 10th of 21Claude Sonnet 5.5 4th of 7Muse Wonder $4.50 blended, 85th percentilecheapest: Qwen3.5-9B sets this group
Models from developers based in the US 36th of 92Claude Opus 5.5 16th of 21Claude Fable 5.2 $4.50 blended, 74th percentilecheapest: Gemma 3 4B 19th of 33Nemotron 3.5 Super VL

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.

Measured cost per task

API cost recorded in our benchmark runs, excluding judging.

TaskBenchmarkCost per task
Product listingCatalogBench≈ $0.035
DeckDeckBench≈ $0.136
Survey plan and analysisSurveyBench≈ $0.333
Game of blitz chessBulletBench≈ $0.207

Cheapest model that passes

On our benchmarks that define a pass for each task: what one passing answer cost in our run, failed attempts included, and whether a cheaper model passed.

  • CatalogBench · cost per 1,000 accepted listings
    Gemini 3.1 Pro Preview: $60.18 (99 of 168 attempts), 22 of 25. Cheapest: GPT-6 Luna at $0.70 (114 of 168 attempts). It passed at least as often.
  • DeckBench · cost per presentable deck
    Gemini 3.1 Pro Preview: $0.81 (1 of 6 tasks), 8 of 12. Cheapest: GPT-6.1 Terra at $0.13 (3 of 6 tasks). It passed at least as often.
  • SurveyBench · cost per sound survey task
    Gemini 3.1 Pro Preview: no sound survey tasks (0 of 9 tasks). Cheapest: GPT-6 Luna at $0.03 (3 of 9 tasks).

Cost per attempt divided by the share that passed, at the run date's prices, excluding judging. Each model at its cheapest setting per passing answer. With a handful of tasks, a pass or two either way is within chance. No estimate where a run had missing outputs, because their cost may not have been recorded.

How good is it, and for what?

Each use case ranks the available models its benchmarks measured, then those priced $3–10 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 16th of 32 available · 9th of 14 at its price
BenchmarkGemini 3.1 Pro PreviewRank among availableBest available
CatalogBenchCatalogBench: reliably publish-ready 44.6% 16th of 32 GPT-6 Astra 75.0%
CatalogBenchCatalogBench (sales brief): reliably publish-ready 16.1% 15th of 32 GPT-6 Astra 69.6%
Decks from an analysisTurning a finished analysis into a deck you could present as it is 27th of 32 available · 14th of 15 at its price
BenchmarkGemini 3.1 Pro PreviewRank among availableBest available
DeckBenchDeckBench: deck rating 732 27th of 32 Qwen4-Max 1,585
User surveysPlanning a user survey and reading its results without being misled 18th of 23 available · 8th of 10 at its price
BenchmarkGemini 3.1 Pro PreviewRank among availableBest available
SurveyBenchSurveyBench: analysis rating 758 18th of 23 Claude Fable 5.2 1,641
Marketing planningPlanning a year of ad spend without overspending 18th of 32 available · 10th of 14 at its price
BenchmarkGemini 3.1 Pro PreviewRank among availableBest available
ROASBenchROASBench: overall score 43.5 18th of 32 Claude Fable 5.2 61.3
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 66th of 159 available · 29th of 35 at its price
BenchmarkGemini 3.1 Pro PreviewRank among availableBest available
OpenHands IndexOpenHands Index: average · reported 60.6% 9th of 31 Claude Fable 5 81.0%
Artificial AnalysisArtificial Analysis: Terminal-Bench 4.0 · reported 4.0% 48th of 107 Kimi K3.1 Pro 67.2%
Artificial AnalysisArtificial Analysis: Terminal-Bench 2.1 · reported 73.8% 34th of 107 Claude Fable 5.1 91.4%
Artificial AnalysisArtificial Analysis: τ-bench banking · reported 21.4% 43rd of 104 Grok 4.6 50.7%
Vending-Bench 2Vending-Bench 2: money after a year · reported $911.21 42nd of 53 GPT-6 Astra $15,515
  • tau2-bench: has not measured this model.
  • 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 36th of 51 available · 19th of 23 at its price
BenchmarkGemini 3.1 Pro PreviewRank among availableBest available
APEX-AgentsAPEX-Agents: tasks passed · reported 35.3% 33rd of 37 Claude Sonnet 5.5 75.5%
GDP.pdfGDP.pdf: professional document tasks · reported 17.0% 20th of 31 GPT-5.6 Sol 30.7%
  • Remote Labor Index: has not measured this model.
Reasoning and knowledgeHard questions across science, maths and general knowledge 29th of 187 available · 17th of 39 at its price
BenchmarkGemini 3.1 Pro PreviewRank among availableBest available
Artificial AnalysisArtificial Analysis: Artificial Analysis Intelligence Index · reported 29.7 54th of 185 Claude Opus 5.5 57.6
Artificial AnalysisArtificial Analysis: Humanity's Last Exam · reported 47.0% 19th of 186 Claude Opus 5.5 61.4%
Artificial AnalysisArtificial Analysis: GPQA Diamond · reported 94.1% 5th of 168 GPT-6 Astra 96.3%
WritingWhat people prefer in blind comparisons, and judged writing quality 9th of 169 available · 5th of 34 at its price
BenchmarkGemini 3.1 Pro PreviewRank among availableBest available
Arena (formerly LMArena)Arena Text: overall · reported 1,487 14th of 153 Gemini 4 Argon 1,525
UGI LeaderboardUGI: writing score · reported 72.2 5th of 124 Gemini 3.8 Flash 78.6
Sticking to the factsSummarising without inventing things, and factual answers 33rd of 139 available · 9th of 31 at its price
BenchmarkGemini 3.1 Pro PreviewRank among availableBest available
Vectara Hallucination LeaderboardVectara: hallucination rate · reported 10.4% 42nd of 71 GPT-5.4 nano 3.1%
Arena (formerly LMArena)Arena Text factuality: overall · reported 1,472 18th of 109 Gemini 4 Argon 1,511
SimpleQA Verified (Epoch AI)SimpleQA Verified: correct answers · reported 73.5% 3rd of 68 GPT-6 Astra 75.6%
Speed under pressureGood decisions against a real clock (fast chess) 5th of 35 available · 3rd of 14 at its price
BenchmarkGemini 3.1 Pro PreviewRank among availableBest available
BulletBenchBulletBench Bullet 60s: ladder Elo 529 9th of 33 Gemini 3.8 Flash 1,068

Nearest alternatives

Against Gemini 3.1 Pro Preview at $4.50 blended, Intelligence Index 29.7. "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 $4.50 blended per million tokens. Cheapest first.

HostQuantisationInputOutputBlended
Googlenot stated$2.00$12.00$4.50
Google AI Studionot stated$2.00$12.00$4.50

Other service tiers, not counted above: Google flex $1.00 in, $6.00 out; Google AI Studio flex $1.00 in, $6.00 out; Google priority $3.60 in, $21.60 out; Google AI Studio priority $3.60 in, $21.60 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 3.1 Pro PreviewClaude Opus 5.5
most intelligent at $3–10 per million tokens
Nemotron 3.5 Super VL
cheapest with similar intelligence
Claude Fable 5.2
highest-scoring overall available today
Price per million tokens $4.50 $8.00$0.24$16.00
Tokens a second 130 96411–
Intelligence Index 29.7 57.626.5–
Spring Prompt overall 46 82–90
Rank among available models, by use case
Product listings 16th of 32 13th–5th
Decks from an analysis 27th of 32 7th–4th
User surveys 18th of 23 4th–1st
Marketing planning 18th of 32 9th–1st
Agents and tool use 66th of 159 3rd132nd–
Professional work 36th of 51 3rd––
Reasoning and knowledge 29th of 187 1st74th–
Writing 9th of 169 5th––
Sticking to the facts 33rd of 139 3rd––
Speed under pressure 5th of 35 ––8th

Shaded figures are better than Gemini 3.1 Pro Preview's. A dash means no result.

Research and examples

Where it has been measured

24th most intelligent of the 38 models priced $3–10 per million tokens. Among models available today, in the top quarter (by percentile) for writing, and reasoning and knowledge; in the bottom quarter for decks from an analysis, user surveys and professional work.

Every result

138 published results from 15 sources, each in the source's own units, with the configuration that produced it.

BulletBench · measured by Spring Prompt · 10 results
MeasureValueRankConfigurationDated
BulletBench Blitz 3+2: cost per gameUS dollars, lower is better $0.21 17 of 22 Gemini 3.1 Pro Preview
provider default reasoning
29 Sep 2026
BulletBench Blitz 3+2: games lost on time% of games, lower is better 50.0% 14 of 22 Gemini 3.1 Pro Preview
provider default reasoning
29 Sep 2026
BulletBench Blitz 3+2: invalid moves% of moves, lower is better 0.0% 1 of 22 Gemini 3.1 Pro Preview
provider default reasoning
29 Sep 2026
BulletBench Blitz 3+2: ladder Eloladder Elo, higher is better 683
487–1,130
9 of 22 Gemini 3.1 Pro Preview
provider default reasoning
29 Sep 2026
BulletBench Blitz 3+2: median move timemilliseconds, lower is better 6.2 s 15 of 22 Gemini 3.1 Pro Preview
provider default reasoning
29 Sep 2026
BulletBench Bullet 60s: cost per gameUS dollars, lower is better $0.0446 61 of 71 Gemini 3.1 Pro Preview (low reasoning)
low reasoning
1 Oct 2026
BulletBench Bullet 60s: games lost on time% of games, lower is better 58.3% 47 of 71 Gemini 3.1 Pro Preview (low reasoning)
low reasoning
1 Oct 2026
BulletBench Bullet 60s: invalid moves% of moves, lower is better 0.0% 1 of 71 Gemini 3.1 Pro Preview (low reasoning)
low reasoning
1 Oct 2026
BulletBench Bullet 60s: ladder Eloladder Elo, higher is better 529
329–698
27 of 71 Gemini 3.1 Pro Preview (low reasoning)
low reasoning
1 Oct 2026
BulletBench Bullet 60s: median move timemilliseconds, lower is better 3.1 s 62 of 71 Gemini 3.1 Pro Preview (low reasoning)
low reasoning
1 Oct 2026

Not shown: Not your cost: prices are those charged on the run date.

CatalogBench · measured by Spring Prompt · 31 results
MeasureValueRankConfigurationDated
CatalogBench (sales brief): channel rules broken% of products, lower is better 0.0% 1 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): claims to check% of products, lower is better 10.1% 28 of 32 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): content quality% of checks, higher is better 97.2% 17 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): failed outputs% of products, lower is better 0.0% 1 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): missing UK information% of products, lower is better 0.0% 1 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): not findable% of products, lower is better 6.0% 26 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): publish-ready listings% of products, higher is better 30.4% 18 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): reliably publish-ready% of products, higher is better 16.1% 15 of 34 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): unsupported claims% of products, lower is better 67.3% 22 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): unsupported claimsclaims per product, lower is better 2.13 21 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): wrong attributes% of products, lower is better 16.1% 16 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): wrong category or variant% of products, lower is better 1.8% 6 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench: channel compliance% of products, higher is better 99.4% 26 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench: channel rules broken% of products, lower is better 0.6% 20 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench: claims to check% of products, lower is better 1.2% 25 of 32 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench: conflicts caught% of conflicts, higher is better 100.0% 1 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench: content quality% of checks, higher is better 96.9% 19 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench: cost per productUS dollars, lower is better $0.0355 28 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench: decision accuracy% of decisions, higher is better 96.6% 18 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench: failed outputs% of products, lower is better 0.0% 1 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench: field accuracy% of missing fields, higher is better 95.4% 4 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench: invented values% of filled values, lower is better 1.5% 15 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench: missing UK information% of products, lower is better 0.0% 1 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench: not findable% of products, lower is better 7.7% 27 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench: publish-ready listings% of products, higher is better 58.9% 19 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench: reliably publish-ready% of products, higher is better 44.6% 16 of 34 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench: time per productseconds, lower is better 22.2 23 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench: unsupported claims% of products, lower is better 26.8% 22 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench: unsupported claimsclaims per product, lower is better 0.66 19 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench: wrong attributes% of products, lower is better 14.9% 9 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026
CatalogBench: wrong category or variant% of products, lower is better 1.8% 6 of 35 Gemini 3.1 Pro Preview
provider default reasoning
30 Sep 2026

Not shown: Not your catalogue: invented products with generated images, so a model's result on your own feed can differ.

DeckBench · measured by Spring Prompt · 17 results
MeasureValueRankConfigurationDated
DeckBench: accurate decks% of tasks, higher is better 50.0% 15 of 34 Gemini 3.1 Pro Preview
provider default reasoning
6 Oct 2026
DeckBench: caveat dropped% of tasks, lower is better 16.7% 15 of 34 Gemini 3.1 Pro Preview
provider default reasoning
6 Oct 2026
DeckBench: clean layout% of tasks, higher is better 50.0% 14 of 34 Gemini 3.1 Pro Preview
provider default reasoning
6 Oct 2026
DeckBench: cost per deckUS dollars, lower is better $0.14 19 of 34 Gemini 3.1 Pro Preview
provider default reasoning
6 Oct 2026
DeckBench: deck ratingrating, higher is better 732
652–799
29 of 34 Gemini 3.1 Pro Preview
provider default reasoning
6 Oct 2026
DeckBench: design quality% of the maximum, higher is better 51.4% 28 of 34 Gemini 3.1 Pro Preview
provider default reasoning
6 Oct 2026
DeckBench: draft figure quoted% of tasks, lower is better 0.0% 1 of 34 Gemini 3.1 Pro Preview
provider default reasoning
6 Oct 2026
DeckBench: findings missing% of tasks, lower is better 0.0% 1 of 34 Gemini 3.1 Pro Preview
provider default reasoning
6 Oct 2026
DeckBench: head-to-head win rate% of comparisons, higher is better 31.9% 29 of 34 Gemini 3.1 Pro Preview
provider default reasoning
6 Oct 2026
DeckBench: layout defects% of tasks, lower is better 50.0% 14 of 34 Gemini 3.1 Pro Preview
provider default reasoning
6 Oct 2026
DeckBench: misleading metric used% of tasks, lower is better 0.0% 1 of 34 Gemini 3.1 Pro Preview
provider default reasoning
6 Oct 2026
DeckBench: presentable decks% of tasks, higher is better 16.7% 12 of 34 Gemini 3.1 Pro Preview
provider default reasoning
6 Oct 2026
DeckBench: recommendation late or wrong% of tasks, lower is better 50.0% 26 of 34 Gemini 3.1 Pro Preview
provider default reasoning
6 Oct 2026
DeckBench: slides needing work% of tasks, lower is better 83.3% 31 of 34 Gemini 3.1 Pro Preview
provider default reasoning
6 Oct 2026
DeckBench: time per deckseconds, lower is better 78.0 9 of 34 Gemini 3.1 Pro Preview
provider default reasoning
6 Oct 2026
DeckBench: unsupported claims% of tasks, lower is better 0.0% 1 of 34 Gemini 3.1 Pro Preview
provider default reasoning
6 Oct 2026
DeckBench: unsupported numbers% of tasks, lower is better 0.0% 1 of 34 Gemini 3.1 Pro Preview
provider default reasoning
6 Oct 2026

Not shown: Not your analysis or your brand: invented companies, six public tasks, one deck per model per task at the provider's default reasoning setting.

ROASBench · measured by Spring Prompt · 9 results
MeasureValueRankConfigurationDated
ROASBench: audience scorescore out of 100, higher is better 67.5 9 of 35 Gemini 3.1 Pro Preview
provider default reasoning
29 Sep 2026
ROASBench: business scorescore out of 100, higher is better 35.9 21 of 35 Gemini 3.1 Pro Preview
provider default reasoning
29 Sep 2026
ROASBench: consistency scorescore out of 100, higher is better 33.9 10 of 35 Gemini 3.1 Pro Preview
provider default reasoning
29 Sep 2026
ROASBench: contribution profitsimulated US dollars, higher is better $507,736 23 of 35 Gemini 3.1 Pro Preview
provider default reasoning
29 Sep 2026
ROASBench: cost of a runUS dollars, lower is better $0.52 25 of 35 Gemini 3.1 Pro Preview
provider default reasoning
29 Sep 2026
ROASBench: months over budgetmonths of 12, lower is better 0 1 of 35 Gemini 3.1 Pro Preview
provider default reasoning
29 Sep 2026
ROASBench: overall scorescore out of 100, higher is better 43.5 18 of 35 Gemini 3.1 Pro Preview
provider default reasoning
29 Sep 2026
ROASBench: planning scorescore out of 100, higher is better 54.9 14 of 35 Gemini 3.1 Pro Preview
provider default reasoning
29 Sep 2026
ROASBench: return on ad spendprofit per $1 spent, higher is better $0.42 23 of 35 Gemini 3.1 Pro Preview
provider default reasoning
29 Sep 2026

Not shown: Not real ad performance: the market is a deterministic simulation of one invented skincare brand.

SurveyBench · measured by Spring Prompt · 20 results
MeasureValueRankConfigurationDated
SurveyBench: analysis ratingrating, higher is better 758
700–814
18 of 25 Gemini 3.1 Pro Preview
provider default reasoning
8 Oct 2026
SurveyBench: analysis scorepoints of 100, higher is better 66.4 16 of 25 Gemini 3.1 Pro Preview
provider default reasoning
8 Oct 2026
SurveyBench: cost per taskUS dollars, lower is better $0.33 19 of 24 Gemini 3.1 Pro Preview
provider default reasoning
8 Oct 2026
SurveyBench: head-to-head win rate% of comparisons, higher is better 32.4% 20 of 25 Gemini 3.1 Pro Preview
provider default reasoning
8 Oct 2026
SurveyBench: leading or double-barrelled questions% of tasks, lower is better 44.4% 13 of 25 Gemini 3.1 Pro Preview
provider default reasoning
8 Oct 2026
SurveyBench: mistaken next steps% of tasks, lower is better 33.3% 16 of 25 Gemini 3.1 Pro Preview
provider default reasoning
8 Oct 2026
SurveyBench: numbers right% of answers, higher is better 100.0% 1 of 25 Gemini 3.1 Pro Preview
provider default reasoning
8 Oct 2026
SurveyBench: objectives unanswerable% of tasks, lower is better 0.0% 1 of 25 Gemini 3.1 Pro Preview
provider default reasoning
8 Oct 2026
SurveyBench: precision ignored% of tasks, lower is better 55.6% 20 of 25 Gemini 3.1 Pro Preview
provider default reasoning
8 Oct 2026
SurveyBench: research scorepoints of 100, higher is better 71.3 16 of 25 Gemini 3.1 Pro Preview
provider default reasoning
8 Oct 2026
SurveyBench: sound research% of tasks, higher is better 0.0% 15 of 25 Gemini 3.1 Pro Preview
provider default reasoning
8 Oct 2026
SurveyBench: stakeholder's view presumed% of tasks, lower is better 22.2% 15 of 25 Gemini 3.1 Pro Preview
provider default reasoning
8 Oct 2026
SurveyBench: survey plan scorepoints of 100, higher is better 82.9 20 of 25 Gemini 3.1 Pro Preview
provider default reasoning
8 Oct 2026
SurveyBench: survey structure% of tasks, lower is better 22.2% 11 of 25 Gemini 3.1 Pro Preview
provider default reasoning
8 Oct 2026
SurveyBench: time per taskseconds, lower is better 170 12 of 24 Gemini 3.1 Pro Preview
provider default reasoning
8 Oct 2026
SurveyBench: traps handled% of traps, higher is better 71.0% 20 of 25 Gemini 3.1 Pro Preview
provider default reasoning
8 Oct 2026
SurveyBench: traps missed% of tasks, lower is better 88.9% 16 of 25 Gemini 3.1 Pro Preview
provider default reasoning
8 Oct 2026
SurveyBench: unsupported findings% of tasks, lower is better 77.8% 14 of 25 Gemini 3.1 Pro Preview
provider default reasoning
8 Oct 2026
SurveyBench: wrong numbers% of tasks, lower is better 0.0% 1 of 25 Gemini 3.1 Pro Preview
provider default reasoning
8 Oct 2026
SurveyBench: wrong recommendation% of tasks, lower is better 44.4% 17 of 25 Gemini 3.1 Pro Preview
provider default reasoning
8 Oct 2026

Not shown: Not your customers or your data: invented organisations with simulated respondents, nine public tasks, one run per model per task at the provider's default reasoning setting.

APEX-Agents · reported by APEX-Agents · 1 result
MeasureValueRankConfigurationDated
APEX-Agents: tasks passed% of tasks, higher is better 35.3% 35 of 39 Gemini 3.1 Pro Preview 1 Oct 2026

Not shown: Not your firm's documents or tools; graded by rubric, not by a client.

Arena (formerly LMArena) · reported by Arena (formerly LMArena) · 13 results
MeasureValueRankConfigurationDated
Arena Agent: confirmed task successIPS effect estimate, higher is better -0.10
-0.12–-0.08
44 of 50 Gemini 3.1 Pro Preview (arena agent) 2 Oct 2026
Arena Agent: praise over complaintIPS effect estimate, higher is better -0.01
-0.04–0.01
31 of 50 Gemini 3.1 Pro Preview (arena agent) 2 Oct 2026
Arena Agent: steerabilityIPS effect estimate, higher is better -0.05
-0.06–-0.03
39 of 50 Gemini 3.1 Pro Preview (arena agent) 2 Oct 2026
Arena Agent: tool groundingIPS effect estimate, higher is better -0.00
-0.01–0.01
37 of 50 Gemini 3.1 Pro Preview (arena agent) 2 Oct 2026
Arena Document: overallArena rating, higher is better 1,459
1,454–1,464
27 of 44 Gemini 3.1 Pro Preview 13 Sep 2026
Arena Search: overallArena rating, higher is better 1,208
1,202–1,213
6 of 34 Gemini 3.1 Pro Preview (arena search) 24 Aug 2026
Arena Text factuality: overallArena rating, higher is better 1,472
1,470–1,474
26 of 182 Gemini 3.1 Pro Preview 2 Oct 2026
Arena Text: business, management and financeArena rating, higher is better 1,477
1,472–1,482
28 of 406 Gemini 3.1 Pro Preview 2 Oct 2026
Arena Text: creative writingArena rating, higher is better 1,480
1,475–1,486
11 of 411 Gemini 3.1 Pro Preview 2 Oct 2026
Arena Text: expert promptsArena rating, higher is better 1,509
1,502–1,515
31 of 363 Gemini 3.1 Pro Preview 2 Oct 2026
Arena Text: instruction followingArena rating, higher is better 1,480
1,475–1,484
20 of 413 Gemini 3.1 Pro Preview 2 Oct 2026
Arena Text: overallArena rating, higher is better 1,487
1,484–1,490
18 of 413 Gemini 3.1 Pro Preview 2 Oct 2026
Arena Text: writing, literature and languageArena rating, higher is better 1,480
1,475–1,485
13 of 412 Gemini 3.1 Pro Preview 2 Oct 2026

Not shown: Not a success rate: the estimated effect of choosing this model as the agent's orchestrator, against an even mix of the models tested, inside Arena's own harness.

Artificial Analysis · reported by Artificial Analysis · 14 results
MeasureValueRankConfigurationDated
Artificial Analysis: Artificial Analysis Coding Indexindex score, higher is better 68.8 56 of 187 Gemini 3.1 Pro Preview 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 29.7 120 of 413 Gemini 3.1 Pro Preview 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 94.1% 7 of 360 Gemini 3.1 Pro Preview 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 47.0% 41 of 419 Gemini 3.1 Pro Preview 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 77.1% 15 of 288 Gemini 3.1 Pro Preview 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 82.0% 44 of 407 Gemini 3.1 Pro Preview 20 Oct 2026
Artificial Analysis: Output speedtokens per second, higher is better 130 51 of 137 Gemini 3.1 Pro Preview 20 Oct 2026
Artificial Analysis: SciCode% of problems, higher is better 58.7% 19 of 205 Gemini 3.1 Pro Preview 20 Oct 2026
Artificial Analysis: Terminal-Bench 2.1% of tasks, higher is better 73.8% 68 of 188 Gemini 3.1 Pro Preview 20 Oct 2026
Artificial Analysis: Terminal-Bench 4.0% of tasks, higher is better 4.0% 114 of 204 Gemini 3.1 Pro Preview 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 53.8% 16 of 282 Gemini 3.1 Pro Preview 20 Oct 2026
Artificial Analysis: Time to first answer tokenseconds, lower is better 14.7 74 of 137 Gemini 3.1 Pro Preview 20 Oct 2026
Artificial Analysis: τ-bench banking% of tasks, higher is better 21.4% 83 of 175 Gemini 3.1 Pro Preview 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 95.6% 18 of 286 Gemini 3.1 Pro Preview 20 Oct 2026

Not shown: Not your codebase or tools.

GDP.pdf · reported by GDP.pdf · 2 results
MeasureValueRankConfigurationDated
GDP.pdf: professional document tasks% of rubric, higher is better 17.0% 28 of 49 Gemini 3.1 Pro Preview 6 Jun 2026
GDP.pdf: professional document tasks% of rubric, higher is better 17.0% 28 of 49 Gemini 3.1 Pro 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 60.6% 9 of 34 Gemini 3.1 Pro Preview (openhands) 30 Jun 2026
OpenHands Index: front end (SWE-Bench Multimodal)% resolved, higher is better 44.1% 4 of 34 Gemini 3.1 Pro Preview (openhands) 30 Jun 2026
OpenHands Index: greenfield (Commit0)% resolved, higher is better 25.0% 14 of 34 Gemini 3.1 Pro Preview (openhands) 30 Jun 2026
OpenHands Index: information gathering (GAIA)% resolved, higher is better 81.8% 4 of 34 Gemini 3.1 Pro Preview (openhands) 30 Jun 2026
OpenHands Index: issue resolution (SWE-Bench)% resolved, higher is better 76.8% 6 of 34 Gemini 3.1 Pro Preview (openhands) 30 Jun 2026
OpenHands Index: testing (SWT-Bench)% resolved, higher is better 75.1% 9 of 34 Gemini 3.1 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.

SimpleQA Verified (Epoch AI) · reported by SimpleQA Verified (Epoch AI) · 1 result
MeasureValueRankConfigurationDated
SimpleQA Verified: correct answers% of questions, higher is better 73.5% 3 of 85 Gemini 3.1 Pro Preview (high reasoning) 10 Aug 2026

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

tau2-bench · reported by tau2-bench · 2 results
MeasureValueRankConfigurationDated
tau2 v1.0.1, banking knowledge, alltools retrieval: consistency (pass^4)% of tasks, higher is better 9.3% 17 of 21 Gemini 3.1 Pro Preview (high reasoning, tau2) 5 May 2026
tau2 v1.0.1, banking knowledge, alltools retrieval: task success (pass^1)% of tasks, higher is better 26.0% 15 of 21 Gemini 3.1 Pro Preview (high reasoning, tau2) 5 May 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 20.0% 187 of 379 Gemini 3.1 Pro Preview (high reasoning) 19 Feb 2026
UGI: requested-length error% off the requested word count, lower is better 19.0% 171 of 379 Gemini 3.1 Pro Preview (low reasoning) 19 Feb 2026
UGI: requested-length error% off the requested word count, lower is better 19.0% 171 of 379 Gemini 3.1 Pro Preview (medium reasoning) 19 Feb 2026
UGI: style adherencescore from 0 to 1, higher is better 0.33 258 of 379 Gemini 3.1 Pro Preview (high reasoning) 19 Feb 2026
UGI: style adherencescore from 0 to 1, higher is better 0.34 218 of 379 Gemini 3.1 Pro Preview (low reasoning) 19 Feb 2026
UGI: style adherencescore from 0 to 1, higher is better 0.33 268 of 379 Gemini 3.1 Pro Preview (medium reasoning) 19 Feb 2026
UGI: writing scorescore out of 100, higher is better 68.2 38 of 379 Gemini 3.1 Pro Preview (high reasoning) 19 Feb 2026
UGI: writing scorescore out of 100, higher is better 72.2 12 of 379 Gemini 3.1 Pro Preview (low reasoning) 19 Feb 2026
UGI: writing scorescore out of 100, higher is better 70.2 21 of 379 Gemini 3.1 Pro Preview (medium reasoning) 19 Feb 2026

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 3.1 Pro Preview 22 Sep 2026
Vectara: hallucination rate% of summaries, lower is better 10.4% 61 of 108 Gemini 3.1 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 $911.21 50 of 63 Gemini 3.1 Pro Preview 1 Oct 2026

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

Embed a score badge

A badge for its Spring Prompt Score and each of our benchmarks that ranked Gemini 3.1 Pro Preview, from the current results. Paste the code into a page or README; it links back to where the result comes from.

Gemini 3.1 Pro Preview is #18 of 32 on the Spring Prompt Score (44 of 100, from 5 of 5 of our benchmarks), measured by Spring Prompt Spring Prompt Score · 5 of 5 benchmarks
<a href="https://springprompt.com/score#google-gemini-3-1-pro-preview"><img src="https://springprompt.com/badge/score/google-gemini-3-1-pro-preview.svg" alt="Gemini 3.1 Pro Preview is #18 of 32 on the Spring Prompt Score (44 of 100, from 5 of 5 of our benchmarks), measured by Spring Prompt" height="20"></a>
Gemini 3.1 Pro Preview is #16 of 32 on CatalogBench (Reliably publish-ready, factual brief: 44.6%), measured by Spring Prompt CatalogBench · Reliably publish-ready, factual brief
<a href="https://springprompt.com/benchmarks/catalogbench?focus=google-gemini-3-1-pro-preview"><img src="https://springprompt.com/badge/catalogbench/google-gemini-3-1-pro-preview.svg" alt="Gemini 3.1 Pro Preview is #16 of 32 on CatalogBench (Reliably publish-ready, factual brief: 44.6%), measured by Spring Prompt" height="20"></a>
Gemini 3.1 Pro Preview is #28 of 33 on DeckBench (Deck rating: 732), measured by Spring Prompt DeckBench · Deck rating
<a href="https://springprompt.com/benchmarks/deckbench?focus=google-gemini-3-1-pro-preview"><img src="https://springprompt.com/badge/deckbench/google-gemini-3-1-pro-preview.svg" alt="Gemini 3.1 Pro Preview is #28 of 33 on DeckBench (Deck rating: 732), measured by Spring Prompt" height="20"></a>
Gemini 3.1 Pro Preview is #18 of 24 on SurveyBench (Analysis rating: 758), measured by Spring Prompt SurveyBench · Analysis rating
<a href="https://springprompt.com/benchmarks/surveybench?focus=google-gemini-3-1-pro-preview"><img src="https://springprompt.com/badge/surveybench/google-gemini-3-1-pro-preview.svg" alt="Gemini 3.1 Pro Preview is #18 of 24 on SurveyBench (Analysis rating: 758), measured by Spring Prompt" height="20"></a>
Gemini 3.1 Pro Preview is #18 of 33 on ROASBench (Score: 43.5), measured by Spring Prompt ROASBench · Score
<a href="https://springprompt.com/benchmarks/roasbench?focus=google-gemini-3-1-pro-preview"><img src="https://springprompt.com/badge/roasbench/google-gemini-3-1-pro-preview.svg" alt="Gemini 3.1 Pro Preview is #18 of 33 on ROASBench (Score: 43.5), measured by Spring Prompt" height="20"></a>
Gemini 3.1 Pro Preview is #25 of 59 on BulletBench (Ladder Elo: 529), measured by Spring Prompt BulletBench · Ladder Elo
<a href="https://springprompt.com/benchmarks/bulletbench?focus=google-gemini-3-1-pro-preview"><img src="https://springprompt.com/badge/bulletbench/google-gemini-3-1-pro-preview.svg" alt="Gemini 3.1 Pro Preview is #25 of 59 on BulletBench (Ladder Elo: 529), measured by Spring Prompt" height="20"></a>

Will Gemini 3.1 Pro 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.