Home/Models/Gemma 2B
Google DeepMind logo

Gemma 2B

by Google DeepMind · Released Jan 2024

Open Source
30.8
avg score
Rank #176
Compare
Better than 24% of all models
Context
N/A
Input $/1M
TBD
Output $/1M
TBD
Type
text
License
Open Source
Benchmarks
16 tested
Data updated today
About

Tested on 16 benchmarks with 29.1% average. Top scores: OpenBookQA (71.5%), HellaSwag (61.9%), PIQA (54.6%).

Capabilities
reasoning
12.3
#122 globally
math
12.6
#183 globally
knowledge
36.7
#157 globally
general
21.1
#40 globally
language
26.6
#129 globally
Benchmark Scores
Compare All
Tested on 16 benchmarks · Ranked across 5 categories
Score Distribution (all 233 models)
0255075100
▲ You are here
BBH

BIG-Bench Hard. 23 challenging tasks from BIG-Bench where prior language models fell below average human performance.

13.6
MUSR

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

11.0
GSM8K

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

17.7
MATH Level 5

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

7.4
OpenBookQA

Elementary science questions with access to a small book of core science facts. Tests reasoning beyond memorization.

71.5
HellaSwag

Sentence completion requiring commonsense reasoning about physical and social situations. Tests real-world understanding.

61.9
PIQA

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

54.6
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Specifications
  • Typetext
  • ContextN/A
  • ReleasedJan 2024
  • LicenseOpen Source
  • Statusbenchmark-only
Available On
Google DeepMind logoGoogle DeepMindTBD
Share & Export
Tweet
Gemma 2B is an open-source text AI model by Google DeepMind, released in January 2024. It has an average benchmark score of 30.8.