Welcome Gemma 2 - Google’s new open LLM

What changed

• Models on the Hub - Hugging Face Transformers integration - Integration with Google Cloud & Inference Endpoints - What is Gemma 2? It comes in two sizes, 9 billion and 27 billion parameters with base (pre-trained) and instruction-tuned versions. - Model Merging: Combines two or more LLMs into a single new model Gemma 2 was trained on Google Cloud TPU (27B on v5p, 9B on TPU v4) using JAX and ML Pathways.

Why it matters

A concrete addition to Practical AI: Tools, Models & Frameworks: it changes what's available to builders today rather than being general commentary.

How it compares

Related prior coverage to compare against:

• Welcome Gemma - Google’s new open LLM
• Welcome Gemma 3: Google's all new multimodal, multilingual, long context open LLM
• Welcome Llama 3 - Meta's new open LLM

• Welcome Gemma - Google’s new open LLM similarity 1.00
• Welcome Gemma 3: Google's all new multimodal, multilingual, long context open LLM similarity 0.69
• Welcome Llama 3 - Meta's new open LLM similarity 0.51