Vertex AI Adds New Evaluation Metrics, Model Updates, and Enhanced Fine-tuning
Action Required
Developers can leverage these new features to improve model performance and explore a wider range of generative AI models, potentially reducing costs and increasing application effectiveness.
AI Impact Summary
Google Vertex AI has released a significant update with new evaluation metrics (perplexity, BLEU, ROUGE), enhanced fine-tuning capabilities with PEFT and Axolotl, and expanded model support including Llama 3.1, Gemma 2, PaliGemma 2, and LLaVA Next. This represents a capability expansion for Vertex AI, providing developers with more tools and options for building and deploying generative AI applications. Users should evaluate these new features to improve model performance and explore the updated model support.
Affected Systems
- Date
- Date not specified
- Change type
- capability
- Severity
- high