NVIDIAhas published a blog announcing optimisations for a new open language model,Google’s Gemma: https://blogs.nvidia.com/blog/google-gemma-llm-rtx-ai-pc.
Teamsfrom Google and NVIDIA worked closely together to accelerate the performance ofGemma — Google’s groundbreaking new 2billion- and 7billion-parameter open language model. Its built from the same research andtechnology used to create the Gemini models — with TensorRT-LLM, an open-sourcelibrary for optimising large language model inference, when running on NVIDIAGPUs in the data centre, in the cloud, and on local RTX AI PCswith NVIDIA RTX GPUs.
- NVIDIA, in collaboration with Google, today launched optimisations across all NVIDIA AI platforms, including local RTX AI PCs, for Gemma — Google’s groundbreaking new 2 billion-and 7 billion- parameter open language models.
- Chat With RTX, an NVIDIA tech demo that uses retrieval-augmented generation and NVIDIA TensorRT-LLM software to give users generative AI capabilities on their local, RTX-powered Windows PCs — will add support for Gemma soon.
Teamsfrom Google and NVIDIA worked closely together to accelerate the performance ofGemma — Google’s groundbreaking new 2billion- and 7billion-parameter open language model. Its built from the same research andtechnology used to create the Gemini models — with TensorRT-LLM, an open-sourcelibrary for optimising large language model inference, when running on NVIDIAGPUs in the data centre, in the cloud, and on local RTX AI PCswith NVIDIA RTX GPUs.