Overview
Gemma 2 was released in June 2024. The 9B and 27B models used knowledge distillation and architectural refinements to perform well above their size at launch.
Highlights
- 2B, 9B and 27B sizes
- 8K-token context window
- Distilled training for strong small-model quality
- Wide tool support
Typical uses
Local assistants, classification, fine-tuning and education.
Things to consider
- Short 8K context compared with newer models.
- Gemma terms of use, not an OSI license.
- Download sizes refer to the default 4-bit quantized versions in the Ollama library; full-precision weights are larger.
Specifications are taken from the model card on Hugging Face, the Ollama library and the provider’s announcement. Benchmark results depend on evaluation settings; see how AI benchmarks work.