Anthropic is an AI-based startup focused on inventing tools and software that uses Artificial intelligence. Millions of investors have backed it up and even Google has invested in this startup. Anthropic has launched a new AI-based GenAI tech, the Claude. Not only this, the company has made some bold statements in favor of their new AI model; they claim that it can easily beat Open AI’s GPT-4 and Google Gemini.
Open AI’s GPT-4 is probably one of the most advanced AI models out there. It has undergone many generational upgrades and changes. Gemini, on the other hand, is a first-gen AI model but has gone through rigorous stages of development and testing. And claiming to beat these AI models is for sure not a small thing. The company must have some guts to make it and seems like Anthropic has it.
What’s Anthropic’s GenAI model, the Claude?
Claude 3 is Anthropic’s first generational multi-model GenAI. It is a family of models that consist of Claude 3 Haiku, Claude 3 Sonnet, and Claude 3 Opus. Out of these three, the Opus is the most powerful model. This tool can do multiple tasks like analysing images, texts, and graphs. Apart from this, it can process multiple photos, charts, technical diagrams, drawings from PDFs, and a lot more.
All three models inside it have an upgraded level of analysing and forecasting according to the brand. One good thing about the Claude is that, unlike competitors, it can process up to 20 images or files at the same time. Also, to comply with legal regulations and to ensure the safety of people, the AI model cannot detect or analyse faces from the documents.
However, there’s a big batch. It’s not a fully fledged AI model as of now, because it cannot generate AI artworks like Gemini or GPT. Anthropic says that these models are focused around image analyzation as of right now. We can expect it to get more features and changes in the future.
Claude 3, according to Anthropic, will initially support a context window of 200,000 tokens, or around 150,000 words. A 1-million token context window, or roughly 700,000 words, will be available to a select few users. That is comparable to the recently released Gemini 1.5 Pro model from Google, which similarly provides a context window with up to a million tokens. Opus and Sonnet are currently accessible via the web, Google’s Vertex AI, Amazon’s Bedrock platform, and Anthropic’s dev console and API. Later this year, haiku will be the next.
Leave a Reply