Anthropic has once again made waves in the artificial intelligence space by unveiling its latest innovation, Claude 3. This cutting-edge AI technology marks a significant leap forward, promising to revolutionize how industries operate and interact with artificial intelligence systems.

Related: What is ChatGPT? The AI Language Processing Tool Explained

Source: Cloudbooklet

Anthropic, an AI company founded by a bunch of ex-OpenAI folks, claims that their new AI model family, Claude 3, performs just as well, if not better, than the big shots from Google and OpenAI. Unlike its predecessors, Claude 3 is also multi-modal, meaning it can understand both text inputs and photos.

Anthropic conducted training for the Claude 3 models using a combination of their proprietary datasets, external third-party data, and publicly accessible information available up to August 2023. In a document unveiling the trio of models, the company stated they were trained using Amazon AWS and Google Cloud hardware. Both companies have invested in Anthropic, with Amazon dropping $4 billion into the company. 

Anthropic claims that Claude 3 will field more questions, grasp longer instructions, and be more pinpoint accurate. Claude 3 can understand more context, meaning it can chew through more info. There’s Claude 3 Haiku, Claude 3 Sonnet, and Claude 3 Opus, with Opus being the biggest and “brainiest model” in the mix. Anthropic reports that Opus and Sonnet are now available on and its API, with Haiku coming soon. All three models can tackle a wide range of cognitive tasks, from reasoning to knowledge extraction, content creation, and code generation.

The New Version of Claude will Answer Awkward Questions

Previous Claude was known to dodge some harmless queries, which the company says “hints at a lack of contextual understanding.” The new models are less likely to shy away from addressing requests that adhere to safety rules, akin to rumours about Meta’s plans regarding Llama 3 when it’s released.

Source: Claude 3

Anthropic shares that the Claude 3 models can deliver near-instantaneous results even when tackling dense material like research papers. In a blog post, they assert that Haiku, the smallest version of Claude 3, is “the quickest and most cost-effective model on the market,” capable of digesting a thick research paper with diagrams and charts “in less than three seconds.”

New Chatbot’s Benefits

Anthropic said that Opus has outperformed most models in several comparative tests. It showcased superior performance at a graduate level compared to OpenAI’s GPT-4, scoring 50.4 percent compared to GPT-4’s 35.7 percent in this test. Additionally, it excelled at answering math questions, coding, and understood reasoning better.

Claude 3 boasts a colossal contextual window of 200,000 tokens, meaning it can process and retain a mammoth amount of information simultaneously. Representing a multi-modal AI package, Claude 3 can handle both textual and graphical inputs.

This functionality allows the AI to respond to queries based on visual content like photos or diagrams, enhancing its versatility and user-friendliness. Anthropic’s President, Daniela Amodei, stressed that clients opting for Claude 3 Opus will benefit from its superior cognitive abilities, particularly in tackling intricate tasks like financial analysis with precision.

Read Also:

The pricing structure of Claude 3 involves charging $15 for every 1 million tokens of data processed by Claude 3 Opus, making it a stellar choice for tasks demanding high cognitive complexity. By comparison, other models like OpenAI’s GPT-4 Turbo charge $10 per million tokens.

The new models have also seen significant improvements compared to the previous Claude 2.1 model. Sonnet, the intermediate model, was twice as fast as Claude 2 and Claude 2.1. “It handles tasks requiring quick responses exceptionally well, such as knowledge retrieval or sales automation,” stated Anthropic.

Disclaimer: All materials on this site are for informational purposes only. None of the material should be interpreted as investment advice. Please note that despite the nature of much of the material created and hosted on this website, HODL FM is not a financial reference resource and the opinions of authors and other contributors are their own and should not be taken as financial advice. If you require advice of this sort, HODL FM strongly recommends contacting a qualified industry professional.