← Back to all models

Gemma 2

GoogleLarge Language ModelOpen Source

Google's open-source lightweight model family. Efficient and capable, designed for on-device and resource-constrained deployments.

Abilities

Text GenerationCode GenerationReasoningOn-deviceMultilingual

Use Cases

On-deviceChatbotEducationResearch

Available in Tools

Availability

Ollama
Google AI Studio
Self-hosted

How to Use

Download from Hugging Face or run via Ollama (`ollama run gemma2`). Multiple sizes available from 2B to 27B.

Pros

  • Open source from Google — well-documented and actively maintained
  • Very efficient for its size — strong quality-per-parameter ratio
  • Runs on consumer hardware — 2B variant works even on phones
  • Multiple sizes (2B/9B/27B) for different hardware constraints

Cons

  • Small context window (8K) — very limited for document processing
  • Weaker than larger models on complex tasks
  • Text-only — no vision or multimodal capabilities
  • Superseded by Gemma 3 for most use cases

What to Use It For

star

Perfect For

On-device AI on constrained hardware

2B variant runs on phones and edge devices with minimal resources

Learning and experimenting with LLMs

Well-documented, easy to run locally, and backed by Google research

thumb_up

Good For

Private local inference

All data stays on your machine — complete privacy guaranteed

warning

Not Recommended

Long document processing

8K context is far too short for most real-world document tasks

Try instead: Gemini 2.5 Pro

Complex reasoning

Small model size limits reasoning depth significantly

Try instead: o3

block

Do Not Use For

Production enterprise chatbots

Quality gap with frontier models is too large for customer-facing applications

Try instead: Claude Sonnet 4

Multimodal tasks

Text-only model — cannot process images, audio, or video

Try instead: GPT-4o

Technical Details

PricingFree (open source)
Parameters2B / 9B / 27B
detail.contextWindow8K tokens
LicenseGemma License

Links