Anthropic released a new AI foundation model today: Claude 3.5 Sonnet, the latest iteration of their large language model (LLM) that now includes multimodal capabilities for both language and images.

Benchmark Standard

Claude 3 Opus is widely regarded as one of the top three generative AI foundation models. Many see it as superior to GPT-4 Turbo and competitive with GPT-4o. However, Claude 3.5 Sonnet has set a new standard among Anthropic’s AI models by being a smaller model that delivers higher quality at a lower price point.

The diagram above emphasizes the implications of delivering higher intelligence with a smaller model. Claude 3.5’s Sonnet is “more intelligent” than Claude 3’s Opus, and it achieves this quality while significantly reducing costs.

Addressing the Google Challenge

GPT-4o is clearly a target competitor for the Claude 3.5 model family. However, so are Google Gemini 1.5 and Gemini 1.5 Flash. First, the Gemini models have dethroned Anthropic as leaders in LLM context window size; while Anthropic promoted an industry-leading 200,000 token context window, Google introduced a model that could handle up to 1,000,000 tokens. Less than three months later, Google announced it would have a 2,000,000 token context window available by the end of the year.

Google also established impressive benchmark scores and then with Gemini Pro 1.5 Flash, added high quality combined with low latency and cost. Anthropic had held the position as the key OpenAI foundation model alternative for a year. Suddenly, Google was making credible claims to that position. The higher performance and lower latency of Claude 3.5 Sonnet directly address these advances made in Gemini Pro 1.5 and Flash.

What’s Next?

Anthropic has raised significant funding and secured several customers along the way, establishing itself as a key alternative to OpenAI. Its $18 billion valuation during its most recent round of funding will be hard to justify. However, Anthropic had already established itself as OpenAI’s primary LLM substitute. Claude 3.5 Sonnet is likely to keep Anthropic in the top tier of LLMs.

Beyond performance, Anthropic plans to add new experiences and capabilities that may provide enhanced value for users. This includes developing new modalities and features to support more use cases for businesses, such as integrations with enterprise applications, and exploring features like Memory, which will enable Claude to remember a user’s preferences and interaction history.

Key Takeaways

• Claude 3.5 Sonnet sets new standards in AI model quality and cost with its smaller footprint.
• The introduction of Claude 3.5 represents a step towards surpassing the performance of OpenAI’s GPT-4o, addressing both context window size and benchmark scores.
• Anthropic is positioning itself as a strong alternative to OpenAI in the AI foundation model space, with plans for continued innovation and integration into enterprise applications.