vertex-ai

Introducing Gemini 1.5 Pro: The Next-Generation LLM on Vertex AI

Tal Peretz

Aug 28, 2024 — 2 min read

In February 2024, Google announced the Gemini 1.5 Pro model, a significant leap from its predecessor, the Gemini 1.0 Ultra. Now available for testing on Vertex AI, this next-generation model is designed to offer enhanced performance and efficiency.

Model Characteristics

The Gemini 1.5 Pro stands out with its new Mixture-of-Experts (MoE) architecture, which ensures better performance while being more efficient to train and serve. One of the key highlights is its extensive context window, which supports up to 128,000 tokens natively and can extend up to 1 million tokens for early testers and eventually all users.

This model is also optimized for multimodal capabilities, handling text, images, and videos seamlessly across a wide range of tasks.

Access and Testing

The Gemini 1.5 Pro was initially released as an experimental model on Vertex AI, accessible via the UI without API support. It is now generally available, with features like context caching and provisioned throughput enhancing its usability.

Features and Capabilities

One of the standout features of the Gemini 1.5 Pro is its long-context understanding, enabling it to process extensive data efficiently. The model also supports code execution, allowing developers to run Python code directly within the model through the Gemini API or Google AI Studio. Additionally, context caching improves speed and reduces costs for AI requests involving repetitive content.

Ethics and Safety

Google places a high emphasis on ethics and safety, and the Gemini 1.5 Pro undergoes extensive testing to align with Google's AI Principles and robust safety policies.

Pricing and Quotas

The pricing structure for the Gemini 1.5 Pro includes tiers starting at the standard 128,000 context window, with options to scale up to 1 million tokens. The experimental model comes with a quota of 2 queries per minute (QPM), which cannot be increased.

Future Developments

Looking ahead, Google is focused on optimizing the model to improve latency, reduce computational requirements, and enhance the overall user experience. Additional features and improvements are on the horizon, including better grounding for accuracy and deeper integration with other Google services.

These advancements make the Gemini 1.5 Pro a powerful tool for developers and enterprises, offering unparalleled capabilities in the realm of AI and machine learning.

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