HolyGrAIL

All tasks / Models

Gemini 3.1 Pro (Preview)

Google's Gemini Pro line; its newest version, Gemini 3.1 Pro, is still offered under a preview model code.

  • Google
  • Closed
  • Released 19 Feb 2026

Official page

Where it ranks

  • S
    Chat & assistants
    Models · Cloud · 1,487 on LMArena Text Arena (overall, style control)
  • A
    Search & research
    Models · Cloud · 1,207 on LMArena Search Arena (style control)
  • S
    Writing
    Models · Cloud · 1,481 on LMArena Text Arena: Creative Writing (style control)
  • –
    Documents & OCR
    Models · Cloud · no result on this task’s benchmarks yet
  • C
    Coding
    Models · Cloud · 1,447 on LMArena WebDev / Code Arena (overall)
  • B
    Data analysis
    Models · Cloud · 23.7% on SpreadsheetBench 2 (models in SWE-agent scaffold)
  • S
    Image understanding
    Models · Cloud · 1,280 on LMArena Vision Arena (overall, style control)
  • C
    Vector graphics & logos
    Models · Cloud · 1,284 on Design Arena SVG
  • A
    Transcription
    Models · Cloud · 2.8% on Artificial Analysis Speech to Text: AA-WER v2 (non-streaming)
  • C
    Slides & presentations
    Models · Cloud · 1,063 on Design Arena Python-PPTX Slides

Facts

Vendor
Google
Family
Gemini Pro
Released
19 Feb 2026
Weights
Closed: API and apps only
License
proprietary
API price
$2 in / $12 out per 1M tokens; $0.02 per minute of audio (Artificial Analysis lists $18.15 per 1,000 minutes (Gemini 3.1 Pro Preview (High), provider Google); divided by 1,000.)
Access
Paid only
Last verified
18 Oct 2026

Price from ai.google.dev, 18 Oct 2026.

Benchmark results

BenchmarkResultEntry on the leaderboardAs ofSource
Artificial Analysis Intelligence Index30Gemini 3.1 Pro Preview
Raw value from the page's embedded model data; the site displays it rounded to an integer.
18 Oct 2026artificialanalysis.ai
Artificial Analysis MMMU-Pro (independent evaluation)82.4%Gemini 3.1 Pro Preview
Raw fraction 0.8243 from the model data embedded in https://artificialanalysis.ai/leaderboards/models, converted to percent; rank 20 of 124 entries with an MMMU-Pro result; no last-updated date shown.
19 Oct 2026artificialanalysis.ai
Artificial Analysis Speech to Text: AA-WER v2 (non-streaming)2.8%Gemini 3.1 Pro Preview (High)
AA-WER v2 (non-streaming), provider Google; embedded value 0.0282 converted to percent and rounded; page shows no last-updated date.
18 Oct 2026artificialanalysis.ai
Design Arena HTML Slides1,153Gemini 3.1 Pro Preview
1,979 battles. Read from the site's own API (page renders client-side).
18 Oct 2026designarena.ai
Design Arena Python-PPTX Slides1,063Gemini 3.1 Pro Preview
1,331 battles. Read from the site's own API (page renders client-side).
18 Oct 2026designarena.ai
Design Arena SVG1,284Gemini 3.1 Pro Preview
Design Arena SVG, rank 18, 4620 battles; marked inactive; rows read from POST https://www.designarena.ai/api/leaderboard (arenaType 'models', category 'svg'); names from https://www.designarena.ai/api/registry
19 Oct 2026designarena.ai
EQ-Bench Creative Writing v31,491gemini-3.1-pro-preview
Elo score from the leaderboard data embedded in https://eqbench.com/creative_writing.js (file last modified 2026-09-24); rank 65 of 140; rubric score 16.04.
24 Sep 2026eqbench.com
IDP Leaderboard (v1.5)81.6%Gemini 3.1 Pro
Overall (mean of OlmOCR, OmniDoc and IDP Core columns) on IDP Leaderboard v1.5; page states its ranking is 'As of April 2026'.
18 Oct 2026idp-leaderboard.org
LMArena Search Arena (style control)1,207gemini-3.1-pro-grounding
From LMArena's official leaderboard-dataset (search_style_control, split latest, category overall, published 2026-10-13), which mirrors https://arena.ai/leaderboard/search (site blocks automated fetches). Rank 14 of 42; 151436 votes; 95% CI 1201.8-1211.4.
13 Oct 2026huggingface.co
LMArena Text Arena (overall, style control)1,487Gemini 3.1 Pro Preview
From LMArena's official leaderboard-dataset (text_style_control, split latest, category overall, published 2026-10-15), which mirrors https://arena.ai/leaderboard/text (site blocks automated fetches).
15 Oct 2026huggingface.co
LMArena Text Arena: Creative Writing (style control)1,481Gemini 3.1 Pro Preview
From LMArena's official leaderboard-dataset (text_style_control, split latest, category creative_writing, published 2026-10-15), which mirrors https://arena.ai/leaderboard/text/creative-writing (site blocks automated fetches). Rank 12 of 414; 23619 votes; 95% CI 1475.7-1486.3.
15 Oct 2026huggingface.co
LMArena Vision Arena (overall, style control)1,280gemini-3.1-pro-preview
From LMArena's official leaderboard-dataset (vision_style_control, split latest, category overall, published 2026-10-09), which mirrors https://arena.ai/leaderboard/vision (site blocks automated fetches). Rank 29 of 165; 46320 votes; 95% CI 1274.6-1285.0.
9 Oct 2026huggingface.co
LMArena WebDev / Code Arena (overall)1,447Gemini 3.1 Pro Preview
From LMArena's official leaderboard-dataset (webdev = Code Arena, split latest, category overall, published 2026-10-18), which mirrors https://arena.ai/leaderboard/webdev. Rank 74; 24203 votes; 95% CI 1441.9-1452.0.
18 Oct 2026huggingface.co
ParseBench (models)69.1%Google Gemini 3.1 Pro
Overall (mean of tables, charts, content faithfulness, semantic formatting, visual grounding) from leaderboard.csv in the ParseBench repo (last leaderboard commit 2026-10-09). Default setting; Thinking Low scores 60.82.
9 Oct 2026github.com
SpreadsheetBench 2 (models in SWE-agent scaffold)23.7%Gemini 3.1 Pro
SpreadsheetBench 2 (321 tasks), run in the authors' SWE-agent scaffold; entry date 2026-03-26; marked verified; read from https://spreadsheetbench.github.io/data/leaderboard-v2-full.json.
18 Oct 2026spreadsheetbench.github.io

Where you can use it

Earlier versions

  • Gemini 3 Pro · 18 Nov 2025
  • Gemini 2.5 Pro · 17 Jun 2025

Compare

Alternatives to Gemini 3.1 Pro (Preview)

Sources