NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/Gemma 2 9B
Google DeepMind logo

Gemma 2 9B

by Google DeepMind · Released Jun 2024

Open Source
51.1
avg score
Rank #139
Compare
Better than 55% of all models
Context
8K tokens (~4 books)
Input $/1M
$0.03
Output $/1M
$0.09
Type
text
License
Open Source
Benchmarks
13 tested
Data as of
About

Gemma 2 9B by Google is an advanced, open-source language model that sets a new standard for efficiency and performance in its size class. Designed for a wide variety of...

Tested on 13 benchmarks · BenchGecko score 51.1. Top scores: Chatbot Arena Elo — Overall (1265.0%), GSM8K (84.9%), IFEval (74.4%).

Capabilities
reasoning
9.7
#180 globally
math
31.5
#184 globally
knowledge
36.0
#220 globally
general
42.1
#60 globally
language
74.4
#68 globally
Benchmark Scores
Compare All
Tested on 13 benchmarks · Ranked across 6 categories
Score Distribution (all 312 models)
0255075100
▲ You are here
reasoningCompare reasoning →
MUSR

HuggingFace MuSR (Multi-Step Reasoning). Tests multi-hop reasoning requiring chaining multiple facts together.

9.7·
mathCompare math →
GSM8K

Grade school math word problems. 8,500 problems testing multi-step arithmetic reasoning. A foundational math benchmark.

84.9·
MATH level 5

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

21.0·
MATH Level 5

HuggingFace evaluation of MATH Level 5 problems. Competition math requiring advanced reasoning and proof construction.

19.5·
knowledgeCompare knowledge →
PIQA

Physical Intuition QA. Tests understanding of everyday physical interactions and commonsense physics.

67.4·
MMLU

Massive Multitask Language Understanding. 57 subjects from STEM, humanities, and social sciences. The most widely-cited knowledge benchmark.

62.8·
MMLU-PRO

HuggingFace MMLU-Pro. Harder version of MMLU with 10 answer choices instead of 4 and more challenging questions.

31.9·
Quick compare:Microsoft logovs Phi 3.5 Mini Instructminimax logovs MiniMax M2.7DeepSeek logovs DeepSeek V3.2
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Similar Models
Microsoft logo
Phi 3.5 Mini Instruct
Microsoft
51.1n/a
minimax logo
MiniMax M2.7
minimax
51.2$0.21/1M
DeepSeek logo
DeepSeek V3.2
DeepSeek
51.3$0.28/1M
Links
Info
Google DeepMindPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@Google DeepMind
Source Code
GitHubHugging Face
BenchGecko API
gemma-2-9b-it
Specifications
  • Typetext
  • Context8K tokens (~4 books)
  • ReleasedJun 2024
  • LicenseOpen Source
  • StatusActive
  • Cost / Message~$0.000
Available On
Google DeepMind logoGoogle DeepMind$0.03
Categories
reasoningmathknowledgegenerallanguage
Learn More
transformeropen weightstokens
Share & Export
Tweet
Related Models
Microsoft logoPhi 3.5 Mini Instruct
51.1
minimax logoMiniMax M2.7
51.2
DeepSeek logoDeepSeek V3.2
51.3
anthracite-org logoMagnum v4 72B
50.9
Gemma 2 9B is an open-source text AI model by Google DeepMind, released in June 2024. It has an average benchmark score of 51.1. Context window: 8K tokens.

Related Models

Phi 3.5 Mini Instruct · MicrosoftMiniMax M2.7 · minimaxDeepSeek V3.2 · DeepSeekMagnum v4 72B · anthracite-orgMiniMax M2.5 · minimax

Benchmarks

Chatbot Arena Elo — OverallGSM8KIFEvalPIQAMMLU

Related Pages

Google DeepMind · ProviderGoogle DeepMind · EconomyAll ModelsCompare ModelsPricingDevelopers · API

Key facts · as of 2026-04-14

  • Gemma 2 9B by Google DeepMind. BenchGecko score 51.1, rank 138 of 312 scored models (normalized average of public benchmark scores).
  • List price $0.0300 input · $0.0900 output per 1M tokens (as of 2026-04-14).

How to cite · data as of 2026-04-14

Gemma 2 9B · benchmarks, pricing and providers. BenchGecko, data as of 2026-04-14. https://benchgecko.ai/model/gemma-2-9b-it

Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP