Gemini 3 Pro: benchmarks, price and rank
Gemini 3 Pro from Google is 50th of 123 rated AI models on Spezia's leaderboard. It costs $2 in, $12 out per million tokens.
- 50th of 123 rated models
- Rating 1,344
- Rated on 11 benchmarks
- Closed
- Released Nov 18, 2025
- 1 million tokens of context
- Knows the world to January 2025
Where it is strongest
Not in the top tenth on any benchmark in use.
Where it falls behind
- 16th of 17 on Remote Labor Index
- 18th of 24 on MRCR v2 (8 needles)
- 45th of 69 on ARC-AGI-2
Price: $2 in, $12 out per million tokens.
Its headline results
The benchmarks that count most in the rating, with its place among the models measured the same way, and where each number came from.
- Remote Labor Index1.3%
Real freelance projects that people were paid to do, from design and video to architecture, games and data work. The model gets the same brief and files the freelancer got and has to hand back finished work.
16th of 17. Independent result from Epoch AI.
- Humanity's Last Exam37.5%
About 2,500 questions written by experts in more than a hundred fields to be as hard as they could make them, many needing a diagram or image.
Independent result from Epoch AI.
- Vending-Bench 2$5,478
The model runs a simulated vending machine business for a year on its own: finding suppliers, ordering stock, setting prices and paying the daily fees.
29th of 60. Independent result from Epoch AI.
- SimpleBench76.4%
Multiple-choice puzzles about everyday situations, space, time and people, written to trip up models that pattern-match instead of think. Ordinary adults do well on them.
15th of 68. Independent result from Epoch AI.
- MMMU-Pro81%
College exam questions that need a picture to answer, such as charts, diagrams, maps and chemical structures, with more answer options than the original MMMU and some questions shown only as a screenshot.
11th of 51. The maker's own number, from Google, via LLM Stats.
- CharXiv Reasoning81.4%
Questions about real charts from scientific papers that need reasoning across the chart, not just reading off a value.
24th of 45. The maker's own number, from Google, via LLM Stats.
- ScreenSpot-Pro72.7%
Screenshots of professional software at full resolution, with an instruction to find the exact button or field to use.
8th of 18. The maker's own number, from Google, via LLM Stats.
- MRCR v2 (8 needles)26.3%
A very long conversation full of similar requests, where the model has to find and repeat particular earlier answers, eight of them at once.
18th of 24. The maker's own number, from Google, via LLM Stats.
- LMArena text1,480
People put a question to two anonymous models and vote for the better answer.
16th of 107. Independent result from LMArena.
- LMArena WebDev1,441
People ask two anonymous models to build the same web page or app and vote for the better one.
61st of 98. Independent result from LMArena.
7 more results in its full profile.
Every result with its source, what it costs per task, what jumped with its release, side by side with any other model, and answers to your questions about when to use it.
Models ranked near it
How the rating works, and every benchmark it uses: the model leaderboard.