To a large extent towards the democratization of artificial intelligence, Google introduced GemmaThe family of AI Open Source language models designed to strengthen the position of programmers and researchers around the world. Gemma is a step forward in Google's obligation to provide AI for everyone, following the footsteps of Tensorflow, Bert and T5.
Gemma derives inspiration from siblings, Gemini, who powers the commercial assistant AI Google. The model was built of the same research and technology as Gemini, but with an emphasis on openness and availability.
In the heart of Gemmy lies versatility, with models available in two sizes: Gemma 2B and GEMMA 7B, each equipped with pre -trained and tuned instructional variants. These models retain flexibility to operate directly on local devices, including desktop computers and laptops.
Key functions:
- RAM support: Gemma supports inference and supervises refinement (SFT) in the main framework, such as Jax, Pytorch and Tensorflow through native Keras 3.0.
- Ease of use: Ready to use Colab and Kaggle notebooks, as well as integration with popular tools, such as Hugging Face, Maxtext, Nvidia Nemo and Tensorrt-LLM, make it easier for programmers to start programmers.
- Implementation options: Pre -trained and tuned instructions Gemma models can work on a laptop, workstation or Google Cloud. The implementation options include Vertex AI and Google Kubernetes Engine (GKE).
- Optimized performance: Gemma works extremely well on various AI hardware platforms, including NVIDIA GPU and Google Cloud TPU.
- Programmer's attraction: Google hopes to attract programmers to the Vertex AI cloud platform via Gemma.
The company aims to match the models of open meta meta, emphasizing transparency and privacy. Thus, Gemma is a more available alternative to closed models.
While Gemma means the most important Google release in the field of open AI language models since the appearance of ChatGPT, also emphasizes the company's unwavering involvement in responsible artificial intelligence. Responsible tools for generative AI Google, introduced together with Gemma, enables programmers tools and tips needed to ensure the safe and ethical implementation of the AI application.
Gemma is now available all over the world! Regardless of whether you are an experienced developer or an interesting researcher, test Gemma's capabilities and contributes to the responsible development of AI.
