Google Gemma 2 release on Ollama

TL;DR

Google Gemma 2 is now available on Ollama in three sizes (2 B, 9 B, 27 B) and uses a new architecture that delivers class‑leading performance, allowing the 27 B model to outperform open models more than twice its size.

New architecture and efficiency gains

Gemma 2 introduces a brand‑new model architecture designed specifically for high performance per parameter. The 27 B variant achieves benchmark results that surpass models with over 50 B parameters, establishing a new efficiency baseline for open‑source large language models.

Model sizes and access

Ollama releases three parameter configurations:

  • Gemma 2 2B – smallest footprint, suitable for low‑resource environments. Command: ollama run gemma2:2b.
  • Gemma 2 9B – mid‑range balance of capability and cost. Command: ollama run gemma2.
  • Gemma 2 27B – largest offering, delivering the highest quality while still remaining more efficient than larger competitors. Command: ollama run gemma2:27b.

All three models are hosted in Ollama’s library and can be pulled with a single CLI command.

Integration with popular tooling

Gemma 2 can be used directly from Python through common LLM wrappers.

LangChain

from langchain_community.llms import Ollama
llm = Ollama(model="gemma2")
llm.invoke("Why is the sky blue?")

LlamaIndex

from llama_index.llms.ollama import Ollama
llm = Ollama(model="gemma2")
llm.complete("Why is the sky blue?")

These snippets demonstrate that developers can swap in Gemma 2 with minimal code changes, leveraging existing LangChain or LlamaIndex pipelines.

Implications for the open‑model ecosystem

The release of Gemma 2 narrows the performance gap between open‑source and proprietary models. By delivering higher quality outputs with fewer parameters, Gemma 2 reduces compute costs for inference and fine‑tuning, making advanced LLM capabilities more accessible to researchers and developers who rely on open models.

Gemma 2 is distributed through Ollama; see the official blog post for download links and further details.

Sources

Related

  • Dispatch
    Google Gemma Open LLM Release

    Google has released Gemma, a family of open-access large language models based on Gemini, available in 2B and 7B parameter sizes with base and instruction-tuned variants.

  • Dispatch
    Google Gemma 2 Release Notes

    Google has released Gemma 2, a family of open-weight LLMs available in 9B and 27B parameter sizes, featuring technical advances in distillation and attention mechanisms to improve performance.

  • Dispatch
    Llama 3 Availability on Ollama

    Ollama has released support for Llama 3, providing access to the 8B and 70B parameter models with significant improvements in training data, context length, and token efficiency over Llama 2.

  • Dispatch
    Gemma 3 Release Notes / What's New

    Google has released Gemma 3, a multimodal, multilingual open-weight LLM family ranging from 1B to 27B parameters with context windows up to 128k tokens.

  • Dispatch
    Google releases Gemma 2 2B, ShieldGemma, and Gemma Scope

    Google released Gemma 2 2B, ShieldGemma safety classifiers, and Gemma Scope sparse autoencoders, expanding open‑source LLM capabilities, moderation tools, and interpretability resources.