Скачать или смотреть Advanced LLM Optimization Techniques

Hosted by Paolo Platter, Co-Founder & CTO at Witboost, with expert insights from Antonio Ingargiola, AI and Big Data Engineering Lead at Agile Lab, and Irene Donato, Lead Data Scientist at Agile Lab, this episode dives into advanced strategies for optimizing Large Language Models (LLMs) in multimodal inference tasks involving audio, text, images and video.

💡 What you'll learn:
• Transitioning from online APIs to hosted LLM models
• Selecting smaller, task-specific models for efficiency
• Techniques like fine-tuning, distillation, quantization, and tensor fusion
• Specialized inference servers: Triton and Dynamo
• Managing horizontal scaling with Kubernetes

Enjoy!
-------------
🔗 STAY CONNECTED!
📺 Subscribe to our channel for more exciting content!
📌 Learn more about Agile Lab: https://www.agilelab.it/
🔔 Follow us on LinkedIn:   / agile-lab  

#llm #machinelearning #aioptimization #kubernetes #tritoninference #datascience #podcast #weareagilers #ai