Claude AI 3.5 Sonnet Android App
The newest iteration of Claude AI, which is available on both the web and as an Android app, boasts some impressive improvements over its predecessors. Claude’s text and code-based skills have been upgraded to match those of leading competitors like Google Gemini, OpenAI GPT-4o and Meta’s Llama 3. The new version also offers improved computer vision capabilities that the company says can transcribe text from imperfect images and graphics and understand complex charts and documents.
Using Claude’s visual comprehension abilities, it can interpret science diagrams, decipher charts and graphs, and solve math problems. The company claims that Claude 3.5 Sonnet can understand more than 80% of written English and can converse in multiple languages, including Chinese, Japanese and Korean. It’s even capable of reading and understanding handwriting.
The Claude AI 3.5 Sonnet Android App team has also improved the model’s ability to recognize humor and nuance, resulting in more natural and engaging conversations. In tests, Claude 3.5 Sonnet outperformed GPT-4o on most benchmarks of reasoning and knowledge. It beat the previous model on graduate level reasoning and equaled it on undergraduate level knowledge, coding proficiency and multilingual text-based math. It’s a bit slower on reasoning-based benchmarks than GPT-4o, but still faster than the average person.
Among the most significant upgrades is a new feature called Artifacts that allows users to edit generated content like text documents and code snippets in real time. Claude will open an additional window that displays the result, similar to how chatbots like Gemini and GitHub’s Copilot work with IDEs. The feature will be released later this year and could prove to be a competitive advantage for Claude against the growing number of rival AI assistants that offer code execution and other professional features.
The Claude AI platform is free for individual use, but it can be integrated with other systems and platforms via its API to enable developers to embed its advanced language models. Anthropic, the company behind Claude, is committed to responsible and safe AI design, which includes rigorous testing and training of its models to reduce misuse. It also relies on input from outside subject matter experts to continuously evaluate its products and train classifiers. The company is also working with child safety specialists to identify and address potential risks associated with its models. This approach to ethical design positions Claude well against other AI assistants that have been used to manipulate children online. It’s an approach that we hope other companies will follow, as the AI industry struggles with the ethical challenges of using AI for child exploitation.