ANTHROPIC’S CLAUDE 3.7 SONNET
Gone are the days when we discussed car models or even the ups and downs of the Formula One celebrated drivers. The world post the launch of ChatGPT on the 30th of Nov. 2022 has been a different world. The release of DeepSeek R1 in Jan 2025, shook up the world again, bringing reasoning, cost, nature of training and being open source as the killer combination of the models to come. The tumble DeepSeek R1 gave to the stock market is unheard of; a nature of market impact a tech product has never made. The unilinear AI growth model of OpenAI came under intense scrutiny and challenge.
Gods of AI; Sam Altman,Jensen Huang, Demis Hassabis and the leading AI / IT behemoths are now forced to think. No single AI model or an AI system can claim to be the final one in research and product creation. In the vastly changed scenario, the AI models have now to through a much more serious multifaceted scrutiny, before its claimed benchmarks are accepted and validated by the world. In the case of GPT-4.5, presumably its release got delayed because it was not reaching the company’s expectations, now all of which is referenced on various parameters.
GPT-4.5 thus made no waves, humongous effort and cost notwithstanding. Now it is the turn of Claude 3.7 Sonnet. Anthropic’s new launch is with a mode of reason through complex problems. The model primarily bets on a whole new approach to AI reasoning. The startup makes it fascinating as it claims that Claude 3.7 Sonnet is the first “hybrid reasoning model.” What does this mean? The AI model can switch between quick responses that require less intensive “thinking” and long step-by-step “extended thinking” within a single system. For the record, the two systems should individually test to be qualitatively better in their respective categories, and working in tandem should give a performance without a lag or shortfall in quality as promised.
Undeniably, it is different from other reasoning models simply because reasoning Anthropic believes is “one of the capabilities a frontier model should have, rather than something provided in a separate model.” As per the Business Insider, the model’s ‘extended thinking’ mode outdoes ChatGPT and Grok, but it can overthink. Claude 3.7 Sonnet’s extended thinking mode has strengths, particularly for creative tasks. Its ability to explore multiple ideas, evaluate them and refine the final output is really praiseworthy. The model is claimed to be designed for real world challenges such as the complex coding problems and agentic tasks. The validation has just begun. The jury is out.
THE PROMISE OF AI MODELS IS TURNING OUT TO BE BREATHTAKING.
Sanjay Sahay
Have a nice evening