OVHcloud launches Olala LLM family, first model runs on GPUs with 8GB memory
India, Nov. 19 -- French cloud provider OVHcloud has announced its first family of large language models (LLMs), starting with Olala mini 7A1B, a compact model designed to support agentic artificial intelligence (AI) applications. The model uses a hybrid attention architecture and can run on a graphics processing unit (GPU) with as little as 8GB of memory.
The company will officially launch the model at its OVHcloud AI Summit on November 19, where it plans to showcase its cloud and AI developments.
Olala mini is built on a 7-billion-parameter architecture, with one billion active parameters distributed across 256 experts per layer. It combines Differential-TPA and Mamba-3 attention mechanisms and uses the BF16 numerical format.
Accordi...
Click here to read full article from source
इस लेख के रीप्रिंट को खरीदने या इस प्रकाशन का पूरा फ़ीड प्राप्त करने के लिए, कृपया
हमे संपर्क करें.