Home/Models/Gemma 3 27B
Google DeepMind logo

Gemma 3 27B

by Google DeepMind · Released Mar 2025

Open SourceMultimodal
25.1
avg score
Rank #193
Compare
Better than 17% of all models
Context
131K tokens (~66 books)
Input $/1M
$0.08
Output $/1M
$0.16
Type
multimodal
License
Open Source
Benchmarks
14 tested
Data updated today
About

Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities,...

Tested on 14 benchmarks with 42.2% average. Top scores: Chatbot Arena Elo — Overall (1365.1%), OpenCompass — IFEval (81.0%), Lech Mazur Writing (79.9%).

Capabilities
coding
17.9
#129 globally
math
38.7
#107 globally
knowledge
45.0
#127 globally
language
81.0
#50 globally
Benchmark Scores
Compare All
Tested on 14 benchmarks · Ranked across 5 categories
Score Distribution (all 233 models)
0255075100
▲ You are here
OpenCompass — LiveCodeBenchV6

OpenCompass Live Code Bench v6. Fresh competitive programming problems to evaluate code generation without memorization.

30.8
Aider polyglot

Multi-language code editing from Aider. Tests editing ability across Python, JavaScript, TypeScript, Java, C++, Go, Rust, and more.

4.9
MATH level 5

Competition-level math from AMC, AIME, and olympiad problems. Level 5 is the hardest tier, requiring creative problem-solving.

74.0
OpenCompass — AIME2025

OpenCompass evaluation on AIME 2025 problems. Tests mathematical reasoning on fresh competition problems.

22.4
OTIS Mock AIME 2024-2025

Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.

19.6
Lech Mazur Writing

Writing quality evaluation by Lech Mazur. Tests prose quality, coherence, and stylistic ability.

79.9
OpenCompass — MMLU-Pro

OpenCompass MMLU-Pro evaluation. Harder knowledge test with more answer choices.

67.8
GeoBench

Geography benchmark testing knowledge of world geography, landmarks, borders, and geopolitical facts.

52.0
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Links
Documentation
BenchGecko API
gemma-3-27b-it
Specifications
  • Typemultimodal
  • Context131K tokens (~66 books)
  • ReleasedMar 2025
  • LicenseOpen Source
  • StatusActive
  • Cost / Message~$0.000
Available On
Google DeepMind logoGoogle DeepMind$0.08
Share & Export
Tweet
Gemma 3 27B is an open-source multimodal AI model by Google DeepMind, released in March 2025. It has an average benchmark score of 25.1. Context window: 131K tokens.