All articles tagged with #vertex ai
Google expands Lyria 3 Pro to three-minute AI tracks, enabling prompts for intros, choruses and bridges, and integrates the model across Gemini, Vertex AI, ProducerAI and other Google products, while promising safeguards and watermarking to address impersonation and copyright issues.
Google unveils Gemini 3.1 Pro, claiming a major boost in reasoning to reclaim the AI throne. Independent testing places it at the top of benchmarks, including ARC-AGI-2 (77.1%), GPQA Diamond (94.3%), LiveCodeBench Pro (2887 Elo), SWE-Bench (80.6%), and MMMLU (92.6%), highlighting improved long-horizon thinking and task planning. The model enables functional outputs like vibe-coded SVGs, complex system synthesis, and interactive 3D design, with strong enterprise feedback. Pricing remains $2 per 1M input tokens, plus separate output fees and context caching charges; licensing is via Vertex Studio / Gemini API, and the release is in Preview ahead of general availability.
Google Cloud has launched its new AI video and image-generation models, Veo and Imagen 3, on Vertex AI. Veo, currently in private preview, offers advanced video generation from text or image prompts, while Imagen 3, available next week, provides detailed text-to-image capabilities with editing features. This makes Google the first major cloud provider to offer a video model, enhancing AI workflows in marketing and advertising. The launch intensifies competition with Amazon and Microsoft in the AI space.
Google's Gemini 1.5 Pro, a powerful generative AI model, is now available in public preview on Vertex AI, offering a context window of up to 1 million tokens. This capability allows for tasks such as analyzing code libraries, reasoning across lengthy documents, and holding extensive conversations with chatbots. The model's multilingual and multimodal features enable it to understand images, videos, and audio streams, making it suitable for tasks like transcribing video clips and analyzing media content across different languages. Early users are leveraging its large context window for tasks such as mortgage underwriting, automating metadata tagging on media archives, and generating, explaining, and transforming code.
Google's Gemini 1.5 Pro has been updated to include audio processing capabilities, allowing it to analyze audio files without the need for written transcripts. The model will be available to the public through Google's Vertex AI platform and outperforms the larger Gemini Ultra model. Additionally, Imagen 2, a text-to-image generation model, will introduce new features such as inpainting and outpainting, along with digital watermarking. Google is also previewing a feature to ground AI responses with up-to-date information from Google Search.
Google has made its Gemini AI model available for companies to use through its Vertex AI tool. This integration allows businesses to create their own internal search engines and chatbots without requiring developers. The Gemini-powered capabilities include searching internal information such as company document repositories, enterprise applications, and websites, as well as generating answers and summaries from plain English questions. Google assures customers that they remain in control of their data and that customer data is not used to train the models. Other capabilities rolled out include Imagen 2, which generates logos and image captions, and Duet AI, an AI collaborator on Google Workspace that helps developers code faster. Some Gemini Pro tools are currently free to use until the AI model is rolled out to the general public.
Google has launched Gemini Pro for developers and enterprises through AI Studio and Vertex AI, offering a 32K context window and supporting text input/output. Gemini Pro Vision is also available, accepting text and image input to output text. The tools are free to use with a 60 requests per minute quota, and Google plans to use the data to improve the model. Gemini will be integrated into other developer tools and consumer-facing products in the future. Pricing for Google AI Studio and Vertex AI will be $0.00025 per 1K characters or $0.0025 per image of input.
Google has made its top-of-the-line artificial intelligence program, Gemini Pro, available as a preview version in its AI Studio programming tool and Vertex AI for enterprise users. Gemini is part of Google's AI hyper-computing infrastructure and utilizes the Tensor Processing Unit (TPU) for enhanced performance. The AI Studio allows individuals and small teams to build applications using natural-language prompting, while Vertex AI is designed for enterprise use with access to corporate data sources. Google Cloud also announced the availability of TPU v5p, which offers four times the performance of the previous version. Gemini Pro is one of three versions, with Ultra in private preview and Nano set for release on mobile devices. Additionally, Google introduced Imagen 2, an enhanced text-to-image neural network, available in the Vertex AI feature called Model Garden.
Google has reportedly delayed the launch of its Gemini AI due to inconsistent performance with non-English queries. However, a public preview of Gemini may be on the horizon, as four Google Gemini models have been revealed in the Vertex AI Model Garden. The delay highlights the intense competition in the AI sector, particularly against OpenAI, Microsoft, and Meta. Gemini has fallen short in handling multilingual tasks effectively, which is significant given Google's global market presence. The setback may also impact the enhancement of other Google products like Bard, Assistant, and Docs. Despite the delay, Google remains committed to advancing Gemini and reshaping the landscape of conversational AI.