Gemini 1.5 Pro Now Available |
Written by Kay Ewbank |
Monday, 29 April 2024 |
Google has released Gemini 1.5 Pro with improvements including Native Audio Understanding, System Instructions, and a JSON mode. Google uses the name Gemini both for its conversational chatbot previously known as Bard, and its multimodal large language model (LLM) developed by Google DeepMind. Gemini Pro refers to the LLM. Google Gemini 1.5 Pro is now available in 180 countries in the form of a public preview of the Gemini API that was first announced in February as a private preview. Gemini Pro significantly increases the context window for a publicly available LLM. Until the release of Gemini 1.5 Pro, the largest context window in the world for a publicly available large language model was 200,000 tokens. 1.5 Pro runs up to 1 million tokens consistently, achieving the longest context window of any large-scale foundation model. Gemini 1.5 Pro will come with a 128,000 token context window by default, but the preview version has access to the experimental 1 million token context window. The updated version adds a native audio (speech) understanding capability so it can hear information for the first time. Google says Gemini 1.5 Pro is now able to reason across both image (frames) and audio (speech) for videos uploaded in Google AI Studio, and "we look forward to adding API support for this soon". The update also has a new File API to make it easy to handle files. The preview also adds features including system instructions and a JSON mode to give developers more control over the model's output. The JSON mode lets developers instruct the model to only output JSON objects. This mode enables structured data extraction from text or images. At the moment, it can be used via cURL, with Python SDK support "coming soon". This release also has improvements to function calling. Users can now select modes to limit the model's outputs, improving reliability. Choose text, function call, or just the function itself. The new version also adds support for developers being able to use Google's next generation text embedding model via the Gemini API. The new model, text-embedding-004, (text-embedding-preview-0409 in Vertex AI), achieves a stronger retrieval performance and outperforms existing models with comparable dimensions, on the MTEB benchmarks. Gemini 1.5 Pro is available in public preview now in Google AI Studio. More InformationRelated ArticlesGoogle Rebrands Bard As Gemini With Subscription Google Releases Gemma Open Models Google Adds Code Generation To Bard To be informed about new articles on I Programmer, sign up for our weekly newsletter, subscribe to the RSS feed and follow us on Twitter, Facebook or Linkedin.
Comments
or email your comment to: comments@i-programmer.info |