In a significant move in the rapidly evolving AI landscape, Mira Murati, the former Chief Technology Officer of OpenAI, has unveiled Inkling, a new multimodal language model developed by her startup Thinking Machines. The 975 billion parameter model is already making waves, leading U.S. open-weight models on the Artificial Analysis Intelligence Index, although it still lags behind some of the top Chinese models in certain benchmarks.
Positioning Inkling in the Competitive AI Market
Thinking Machines is positioning Inkling not as the most powerful model on the market, but as a strong base for fine-tuning and customization. This strategic approach reflects a broader industry trend where companies are focusing on building flexible, adaptable AI tools rather than just raw computational power. Inkling is priced at $1.87 per million input tokens, a pricing model that aligns with current market standards for enterprise AI services.
Implications for the Global AI Race
The release of Inkling comes at a time when the global AI race is intensifying, with both U.S. and Chinese firms vying for dominance in open-weight models. While U.S. labs are making strides, particularly in areas like multimodal understanding and open-source collaboration, China's advancements in model size and performance continue to challenge their Western counterparts. Murati’s move signals a new chapter in the AI ecosystem, where former industry leaders are leveraging their expertise to push the boundaries of what’s possible in open AI development.
Conclusion
With Inkling, Thinking Machines is not just introducing a new model but also redefining how open AI can be used in practical applications. As the AI landscape continues to evolve, tools like Inkling may become foundational for developers and enterprises looking to build upon strong, accessible AI infrastructures.
