Gemma 2 9B
Gemma 2 9B IT is a refined, instruction-tuned iteration of Google's Gemma 2 9B model. This enhanced model was developed using a massive dataset of 8 trillion tokens encompassing web content, code, and mathematical data. It incorporates advanced techniques such as sliding window attention, logit soft-capping, and knowledge distillation. Moreover, Gemma 2 9B IT is specifically optimized for dialogue-based applications, achieved through supervised fine-tuning, distillation, reinforcement learning from human feedback (RLHF), and strategic model merging via WARP.
Model Specifications
Technical details and capabilities of Gemma 2 9B
Core Specifications
9.2B Parameters
Model size and complexity
8000.0B Training Tokens
Amount of data used in training
8.2K / 8.2K
Input / Output tokens
June 26, 2024
Release date
Capabilities & License
Performance Insights
Check out how Gemma 2 9B handles various AI tasks through comprehensive benchmark results.
Detailed Benchmarks
Dive deeper into Gemma 2 9B's performance across specific task categories. Expand each section to see detailed metrics and comparisons.
Math
GSM8K
Coding
HumanEval
MBPP
Reasoning
HellaSwag
Knowledge
MMLU
MATH
Non categorized
PIQA
BoolQ
Winogrande
TriviaQA
Providers Pricing Coming Soon
We're working on gathering comprehensive pricing data from all major providers for Gemma 2 9B. Compare costs across platforms to find the best pricing for your use case.
Share your feedback
Hi, I'm Charlie Palars, the founder of Deepranking.ai. I'm always looking for ways to improve the site and make it more useful for you. You can write me through this form or directly through X at @palarsio.
Your feedback helps us improve our service