← Back to all models
G

Gemma 2 9B

by Google DeepMind
Efficient Free & Open Source 🏆 Ranked #69 of 85
61.7
Overall Score
out of 100
About

Google DeepMind's 9B open-source model from the Gemma 2 family, using interleaved local and global attention. Gemma 2 9B competes with models twice its size and is one of the best-performing small open-source models available.

Key Metrics
Context Window
8K
tokens
Avg Response
450
milliseconds
Input Cost
$0.08
per million tokens
Output Cost
$0.08
per million tokens
Arena ELO
1190
Chatbot Arena rating
MT-Bench
8.3
out of 10
Benchmark Scores
MMLU
71.3%
HumanEval
71.0%
MATH
58.0%
GPQA
33.0%
MT-Bench
83.0/10
Capability Profile
Strengths & Limitations
Strengths
✓ Strong for size ✓ Open source ✓ Safety-tuned ✓ Fast ✓ Good reasoning
Limitations
⚠ Limited context (8K) ⚠ Less multilingual ⚠ Below newer Gemma 3
Ideal Use Cases
Personal AI assistants Chatbots Research Code assistance Summarisation
Model Details
Provider Google DeepMind
Released 2024-06-27
Type Free & Open Source
Multimodal No
Tier Efficient
Global rank #69 / 85
Pricing (USD)
Input tokens $0.08/M
Output tokens $0.08/M
Per 1,000 tokens ≈ $0.0001 input / $0.0001 output
All Benchmarks
MMLU 71.3%
HumanEval 71.0%
MATH 58.0%
GPQA 33.0%
MT-Bench 8.3/10
Arena ELO 1190
Compare this model View Rankings

You might also consider