vertex-ai

Vertex AI's Latest Updates: Gemini 1.5 Pro Model and New LLM Capabilities

Tal Peretz

Aug 28, 2024 — 2 min read

Google's Vertex AI platform continues to evolve, with the latest updates introducing powerful enhancements to its large language models (LLMs). These upgrades aim to improve the usability and performance of generative AI applications. Here's a comprehensive look at the recent advancements:

Gemini 1.5 Pro Model

The Gemini 1.5 Pro model is now available in public preview. This model stands out with a 1-million-token context, enabling it to natively reason over vast amounts of data. Additionally, it includes advanced audio processing capabilities, allowing it to handle speech and audio from videos, as well as transcription features.

New Features and Capabilities

Vertex AI has introduced several new features to enhance its LLM capabilities:

Grounding Capabilities: By leveraging Google Search and enterprise data sources with retrieval augmented generation (RAG), Vertex AI reduces hallucinations and improves response accuracy.
Prompt Management: New services for organizing, tracking, and modifying prompts streamline the process of creating, editing, and managing prompts for machine learning models.
Audio Processing: The Gemini 1.5 Pro model's ability to process audio streams, including speech and audio from videos, enables sophisticated cross-modal analysis.

Other Models and Updates

In addition to the Gemini 1.5 Pro, Vertex AI has introduced several other models and updates:

Imagen 2: This model can generate four-second live images from text prompts and features new image editing capabilities, supporting generative AI applications.
Llama 3.1: Available in preview, the Llama 3.1 405B model offers synthetic data generation, model distillation, and multilingual translation.
Anthropic Claude 3.0 Opus: Now generally available, this model is accessible through Vertex AI.
Model Garden: Vertex AI's platform now includes a wide range of models, including third-party options like Meta’s Llama 2 and Anthropic’s Claude 2, providing greater choice and customization for users.

Regional APIs and Model Parameters

Generative AI on Vertex AI regional APIs are now available in regions such as us-east5, me-central1, and me-central2. Additionally, Gemini models now support parameters like frequencyPenalty and presencePenalty, enabling users to control response diversity and repetition.

These updates underscore Google's commitment to enhancing the capabilities and usability of its Vertex AI platform, empowering users to build more effective and accurate generative AI applications.

Unleashing Creativity with Stability.SD3-Large-V1:0 on Amazon Bedrock

The advent of Stability.SD3-Large-V1:0, a revolutionary text-to-image model, marks a significant milestone in the realm of generative AI. Integrated into Amazon Bedrock, this model is designed to transform how digital assets are created, offering unparalleled performance and quality in image generation. Model Architecture and Performance At the heart

Unlocking the Potential of Bedrock/Meta-Textgeneration-Llama-2-7B with AWS SageMaker

The Bedrock/Meta-Textgeneration-Llama-2-7B is a cutting-edge Large Language Model (LLM) developed by Meta as part of the Llama 2 series. This series ranges from 7 billion to 70 billion parameters, with the 7B model offering a balance of size and capability. Its development involved Supervised Fine-Tuning (SFT) and Reinforcement Learning

Exploring Azure AI's Phi-3-Medium-128K-Instruct: A Revolutionary LLM

Azure AI's Phi-3-Medium-128K-Instruct is a breakthrough in the field of language models, bringing forth a dense decoder-only Transformer model with 14 billion parameters. This model, part of Microsoft's Phi-3 family, stands out for its impressive performance across a wide range of benchmarks, making it a go-to

Unlocking the Potential of Azure AI's Phi-3.5-MoE-Instruct: A New Era in Artificial Intelligence

The Azure AI Phi-3.5-MoE-Instruct model, a groundbreaking addition to Microsoft's Phi-3.5 series, is revolutionizing the landscape of artificial intelligence with its innovative architecture and impressive capabilities. Model Architecture At the heart of the Phi-3.5-MoE-Instruct is the Mixture-of-Experts (MoE) model architecture, which includes 16 experts, each