Skip to content
AI Model Radar

gemma 7b

google/gemma-7b
Open Source · chat · open-weights
GA Alert me on changes
Context
—
Max output
—
Input $/1M
$0.07
Output $/1M
$0.07
Modalities
text
Released
08 Feb 2024
License: gemma · google/gemma-7b
AI summary
● machine-written

Google Gemma 7B open-weights text model released February 2024

Gemma 7B is Google's open-weights language model designed for text generation tasks. The model is available under the Gemma license with open access to weights. It has been adopted across multiple inference providers and sees active use in production deployments.

What's new
  • Open-weights model with manual gating on Hugging Face
  • Available on inference providers including Featherless AI
  • Achieves 33.73 score on MMLU-Pro benchmark
  • Compatible with text-generation-inference and Inference Endpoints
Best for
Text generation tasksOpen-source deployment scenariosFine-tuning applications via QLoRACost-sensitive inference workloads
Sources

Source: https://huggingface.co/google/gemma-7b