OverviewAll LeaderboardsMethodologyComing soon
New Chat
Coding IndexGemini 3.1 Pro Preview
Google

Gemini 3.1 Pro Preview

Gemini 3.1 Pro Preview is a proprietary model from Google. It ranks #54 of 65 models on the Coding Index with 990. Its best signal rank is #40 of 65, on Bash Reliability.

Coding Index Overview

Rank
#54

Rank spread 48 – 59.

Coding Index
990± 28
Pareto frontier

237 points under the frontier

Gemini 3.1 Pro Preview
Developer
Google
License
Proprietary
Token price (in / out)
$1 / $6 per 1M
Sessions
12,938
Ranking updated
Nov 13, 2026

Cost, Time, and Tokens

Cost per task$1.23
GPT-6 Luna (Low)$0.01AnthropicClaude Fable 5.2 (Max)$18.90

The median is $1.15.

Time per task4.9 min
MetaMuse Ember (Max)45 sMiMo-V2.6-Flash43.2 min

The median is 11.8 min.

Output tokens per task17.5K
GPT-6 Luna (Low)6.5KAnthropicClaude Sonnet 5.5 (Max)264.8K

The median is 74.9K.

Coding Ranking

8009001,000
  1. 1,038
    Qwen3.8-27B#47
  2. 1,031
    Gemini 3.7 FlashHigh#48
  3. 1,031
    GPT-6 LunaLow#49
  4. 1,021
    Mistral Large 4#50
  5. 1,021
    Tencent
    Hy3#51
  6. 993
    Reflection AI
    Beam 1.0 501B A23B#52
  7. 990
    Qwen3.7-Max#53
  8. 990
    Gemini 3.1 Pro Preview#54
  9. 982
    Synlan
    Spark-X2.6#55
  10. 973
    Meta
    Muse Spark 1.2xHigh#56
  11. 971
    MiniMax-M3#57
  12. 964
    Qwen3.7-Plus#58
  13. 960
    Meta
    Muse Spark 1.1#59
  14. 912
    Mistral Medium 3.5#60
  15. 908
    NVIDIA-Nemotron-3.5-Super-VL-120B-A12BNVFP4#61

Signal Performance

Gemini 3.1 Pro PreviewLeader
Rank on the board
  • Task CompletionHow often the model gets users to confirm the task is done.
    Claude Sonnet 5.5 (Max) · 1,356Claude Sonnet 5.5 (Max) · 1,356Gemini 3.1 Pro Preview · 997
    #56of 65
  • Praise vs ComplaintHow often the model earns more explicitly positive responses than negative ones.
    Claude Fable 5.2 · 2,017Claude Fable 5.2 · 2,017Gemini 3.1 Pro Preview · 1,012
    #48of 65
  • SteerabilityHow well the model lands user corrections when they push back.
    Claude Fable 5.2 · 1,454Claude Fable 5.2 · 1,454Gemini 3.1 Pro Preview · 907
    #58of 65
  • Bash ReliabilityHow often the model’s bash commands succeed. Errors outside its control aren’t counted. E.g. network issues, permission denied, etc.
    Claude Sonnet 5 · 1,234Claude Sonnet 5 · 1,234Gemini 3.1 Pro Preview · 1,097
    #40of 65
9001,0001,1001,2001,3001,4001,5001,6001,7001,8001,9002,0002,100

USE CASES

  • Chat with AI
  • Build Apps & Websites
  • Write & Edit Text
  • Search the Web
  • Generate Images
  • Generate Videos
  • Chose any model
  • Compare Models Side by Side

LEADERBOARD RANKINGS

  • Overall
  • Agent
  • Alignment Index
  • Text
  • WebDev
  • Image-to-WebDev
  • Text to Image
  • Image Edit
  • Text to Video
  • Image to Video
  • Video Edit
  • Vision
  • Document
  • Search

COMPANY

  • About Us
  • How It Works
  • Blog
  • Careers
  • Leaderboard Changelog
  • Product Changelog
  • Help Center
  • FAQ

LEGAL

  • Terms
  • Privacy
  • Cookies

FOLLOW

  • X
  • LinkedIn
  • YouTube
  • Discord

© Arena Intelligence