Home/Models/Gemma 2 27B

Gemma 2 27B

Open Source
by Google DeepMind· Released 2024-06
Official Docs ↗

Gemma 2 27B is Google DeepMind's open-weight model optimized for local deployment. It achieves impressive benchmark scores for its size and runs comfortably on consumer GPUs, making it a popular choice for private, on-premise AI deployments.

Input Price
$0.27/1M
per 1M tokens
Output Price
$0.27/1M
per 1M tokens
Context Window
8K tokens
max tokens
ELO Score
1220
LMSYS Arena

Pricing Breakdown

VolumeInput CostOutput CostCombined (50/50)
1,000 tokens$0.0003$0.0003$0.0003
10,000 tokens$0.0027$0.0027$0.0027
100,000 tokens$0.0270$0.0270$0.0270
1,000,000 tokens$0.2700$0.2700$0.2700

Strengths

  • Best open-source model at 27B scale
  • Runs on consumer hardware (RTX 3090)
  • Strong instruction following
  • Good for fine-tuning

Weaknesses

  • Very small 8K context window
  • Not suitable for long document tasks

Best For

codingsummarizationtranslationlow cost

Compare Gemma 2 27B With

Gemma 2 27B vs o3
OpenAI
Gemma 2 27B vs DeepSeek R1
DeepSeek
Gemma 2 27B vs o1
OpenAI
Gemma 2 27B vs Gemini 2.0 Flash
Google
Gemma 2 27B vs DeepSeek V3
DeepSeek
Gemma 2 27B vs Claude Sonnet 4.6
Anthropic

Frequently Asked Questions

How much does Gemma 2 27B cost?

Gemma 2 27B costs $0.27/1M per 1M input tokens and $0.27/1M per 1M output tokens.

What is Gemma 2 27B's context window?

Gemma 2 27B has a context window of 8K tokens, which means it can process up to 8,192 tokens in a single request.

Is Gemma 2 27B open source?

Yes, Gemma 2 27B is open source. The model weights are publicly available and can be self-hosted.

What is Gemma 2 27B best used for?

Gemma 2 27B is best suited for: coding, summarization, translation, low-cost. Best open-source model at 27B scale.

What is Gemma 2 27B's ELO score?

Gemma 2 27B has an ELO score of 1220 on the LMSYS Chatbot Arena leaderboard, placing it in the mid-tier range.