Compare Start here

Google DeepMindLanguage modelSuperseded

Gemini 3 Pro

Leapfrogged the field in November 2025 and topped LMArena on launch day.

Compare head-to-head Race it in 3D

Overall Intelligence

#22

of 56
9.0 pts behind #1

0

races led

323

weeks at #1

79.7

score

Released: Nov 18, 2025; 11 mo ago
Context: 1M; 65K max output
Price: $2 / $12; per 1M in/out · $4.50 blended
Speed: 105 tok/s; 8s to first token
Modalities: text · image · audio · video; outputs text
Weights: Closed; API and apps

The 30-second version

Google's November 2025 flagship that briefly took the overall lead and pushed multimodal reasoning forward.

Key takeaways

1Led several reasoning races at launch.
2Big step up in agentic and coding skills for Gemini.

Where it ranks

Current snapshot · illustrative scores · click any race to watch it

Gemini 3 ProBest on each axis

Overall

22Overall+9.0 pts

Coding

28Coding+13.1 pts 19Frontend & UI+57 Elo 32Backend & APIs+13.8 pts 26Distributed Systems+14.6 pts 23Full-Stack Apps+12.5 pts 23Debugging & Repair+13.0 pts 30Agentic Coding+16.8 pts 24Code Review & Security+11.6 pts 22Data Science & SQL+8.8 pts 27Mobile Apps+11.0 pts 32Terminal & DevOps+19.2 pts 25Competitive Programming+340

Hard Reasoning

21Hard Reasoning+10.9 pts 21Frontier Questions+19.1 pts 18Abstract & Novel Reasoning+14.0 pts 21Logic & Puzzles+8.2 pts 30Planning & Strategy+13.1 pts

Math

22Math+11.4 pts 20Competition Math+7.9 pts 22Research Math+15.7 pts 23Applied & Quantitative+10.1 pts

Science

22Science+11.1 pts 22Graduate Science QA+6.8 pts 23Biology & Medicine+10.2 pts 21Physics & Chemistry+8.8 pts 23Research Assistance+10.8 pts

Writing & Language

24Writing & Language+11.6 pts 24Creative Writing+76 Elo 27Technical Writing+10.2 pts 27Editing & Style+9.5 pts 21Multilingual+6.0 pts 28Instruction Following+9.7 pts 27Conversation & EQ+73 Elo

Multimodal

7Multimodal+9.0 pts 12Image Understanding+7.3 pts 10Documents & Charts+7.2 pts 4Video Understanding+8.4 pts 8Speech & Audio+7.7 pts

Tool Use & Agents

25Tool Use & Agents+13.1 pts 27Tool Use+10.7 pts 22Computer Use+17.1 pts 27Web Research+16.9 pts 27Long-Horizon+23.1 hrs

Long Context

16Long Context+8.9 pts 171M Recall+7.8 pts 16Long Reasoning+7.5 pts

Speed & Cost

36Speed & Cost+44.3 pts 20Output Speed+418 tok/s 48Latency+7.67s 32Cost-Efficiency+46.6 pts

Trust & Factuality

30Trust & Factuality+13.7 pts 30Factuality+11.6 pts 26Honesty & Calibration+11.1 pts

Overall Intelligence over time

#1 frontierGemini 3 ProHaiku 5.5Qwen3.6-Max

Update history

The distilled version first; open any release for the deeper breakdown.

The Gemini Pro family

2.5 ProMarch 2025 Gemini 3 ProNovember 2025 3.1 ProFebruary 2026 Gemini 4 ArgonOctober 2026

60-second briefing

Gemini 3 Pro, explained

AI video summary · placeholder render from the auto-written script

From Google's own guide, distilled

Gemini prompt design strategies

Structure prompts as persona, task, context and format; the format line is the most often forgotten.
Few-shot examples set the pattern; keep their formatting identical so the model copies structure, not noise.
Gemini reads images, video and PDFs natively, so attach the source instead of describing it.
Place media before the text instruction that refers to it.
Break complex jobs into a chain of smaller prompts and pass outputs forward.

All official guides, distilled →

About Google DeepMind

Google's AI lab behind Gemini, Veo, Imagen and Nano Banana.

Optimising for

Native multimodality (image, video, audio), very long context, science, and tight integration with Search, Workspace and Android.

London, UK / Mountain View, US · founded 2010

Closest rivals

Gemini 3 Pro is 22nd in overall intelligence; leader is Claude Opus 5.5.