Anthropic, a renowned AI business, has unveiled a new model dubbed Claude 3.5 Sonnet, which has established a new benchmark for AI performance. This mid-tier model exceeds the company’s current top-tier model, the Claude 3 Opus, in a variety of tests.
Claude 3.5 Sonnet is available for free via Claude.ai and the Claude iOS app. Users with the Claude Pro and Team subscriptions have higher usage limitations. The model is also available via the Anthropic API, Amazon Bedrock, and Google Cloud’s Vertex AI. The cost is $3 per million input tokens and $15 per million output tokens, and the context window may hold up to 200,000 tokens.
According to Anthropic, Claude 3.5 Sonnet excels in graduate-level reasoning, undergraduate-level knowledge, and coding. It excels in understanding subtleties, comedy, and complicated directions, resulting in high-quality, natural-sounding output.
The model is twice as fast as Claude 3 Opus, making it suited for complicated jobs like as contextual customer service and multi-step processes. In an internal coding test, Claude 3.5 Sonnet answered 64% of the issues, but Claude 3 Opus only solved 38%.
Claude 3.5 Sonnet also has higher visual capabilities, outperforming Claude 3 Opus in conventional vision tests. It is especially effective for activities that need visual thinking, such as reading charts and graphs, and it can reliably transcribe text from defective pictures, which is beneficial in businesses such as retail, logistics, and financial services.
In addition to deploying the model, Anthropic has added a new feature called Artifacts to Claude.AI. This feature enables users to see, edit, and expand on AI-generated material in real time, creating a more collaborative work environment.
Despite its increased capabilities, Claude 3.5 Sonnet maintains Anthropic’s dedication to safety and privacy. To prevent abuse, the model is rigorously tested. External specialists, including the UK AI Safety Institute (UK AISI) and Thorn child safety experts, assisted in testing and improving the model’s safety features.
Anthropic is committed to respecting user privacy. They claim, “We do not train our generative models on user-submitted data unless the user explicitly grants us permission.” To far, we have not trained our generative models with any data given by customers or users.”
Looking ahead, Anthropic intends to offer additional Claude 3.5 models later this year, including the Claude 3.5 Haiku and Claude 3.5 Opus. They are also developing new capabilities to enable additional corporate applications, such as interfaces with enterprise software and a memory function to provide more personalised user experiences.
The AI & Big Data Expo, held in Amsterdam, California, and London, will provide opportunities to learn more about AI and big data from industry professionals. This event is held with other significant events, including the Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo.
Source: Ryan Daws, TechForge Media. Original article published on TechForge Media’s website. You can check out the full article here.
Hi, I'm Voss Xolani, and I'm passionate about all things AI. With many years of experience in the tech industry, I specialize in explaining the functionality and benefits of AI-powered software for both businesses and individual users. My content explores the latest AI tools, offering practical insights on how they can streamline workflows, boost productivity, and drive innovation. I also review new software solutions to help readers understand their features and applications. Beyond that, I stay up-to-date with AI trends and experiment with emerging technologies to provide the most relevant information.