Anthropic, a startup backed by Google and Amazon.com, has announced the release of Claude 3, its latest and most advanced language model to date.

It is available to trial for free, although there is a paid upgrade to a version known as Claude 3 Opus, which is claimed to outperform rival models GPT-4 from OpenAI and Gemini 1.0 Ultra from Google on various benchmark exams.

The family includes three state-of-the-art models in ascending order of capability: Claude 3 Haiku, Claude 3 Sonnet, and Claude 3 Opus. Each successive model offers increasingly powerful performance, allowing users to select the optimal balance of intelligence, speed, and cost for their specific application.

The Claude 3 models can power live customer chats, auto-completions, and data extraction tasks where responses must be immediate and in realtime.

According to Anthropic, the Claude 3 models are better at following complex, multi-step instructions.

“They are particularly adept at adhering to brand voice and response guidelines, and developing customer-facing experiences our users can trust. In addition, the Claude 3 models are better at producing popular structured output in formats like JSON - making it simpler to instruct Claude for use cases like natural language classification and sentiment analysis.

"Claude 3 is a transformative technology that pushes the boundaries of what is possible with large language models," said Dario Amodei, research director at Anthropic. "Through architectural innovations and advanced training techniques, we have created a system that can engage in multi-turn dialogue, answer follow-up questions, and even tackle open-ended tasks with remarkable fluency and coherence."

Some of the key capabilities include:

- Improved reasoning and multi-task performance across a wide range of domains;

- Advanced language understanding enabling natural, contextualized communication;

- Ability to break down complex problems into steps and provide detailed solutions; and

- Strong grasp of factual knowledge from its broad training data.

Anthropic has developed robust AI safety systems alongside Claude 3 to ensure responsible deployment aligned with human interests. However, the company is treading cautiously by rolling it out gradually and closely monitoring its use cases.

With Claude 3, Anthropic aims to unlock new frontiers for artificial intelligence in arenas like open-ended problem solving, creative ideation, tutoring and semantic search.

https://claude.ai/