Thinking Machines Lab Unveils Inkling: A New Open-Source AI Model

Washington DC, United States Jul 15, 2026 ALN: Thinking Machines Lab, an artificial intelligence company founded by former OpenAI members, has officially launched its first model, named Inkling. This new model is open-weight, allowing researchers and startups to download and modify it freely. The emergence of open-source AI models like Inkling represents a significant shift in the landscape of artificial intelligence, as it democratizes access to advanced technology that was previously confined to well-funded corporations.

According to a blog post by the company, Inkling was trained from scratch to interpret audio and video inputs alongside text. This multi-modal capability is particularly noteworthy, as it allows the model to process and understand different types of data simultaneously, which is a crucial aspect of developing more sophisticated AI systems. While it may not top popular benchmarks, it excels in various tasks, demonstrating advanced reasoning and coding capabilities. The model is substantial in size, featuring 975 billion parameters, and requires a cluster of specialized chips for operation. The sheer scale of Inkling places it among the largest AI models currently available, which often leads to improved performance in complex tasks.

In a notable development, the lab utilized Inkling to refine and enhance its own capabilities, showcasing how AI models are increasingly being employed to improve AI technology itself. This self-improvement aspect of AI is an area of growing interest among researchers, as it raises questions about the future of AI development and the potential for machines to autonomously enhance their own functions. The implications of such capabilities could be profound, influencing how AI is integrated into various sectors, from healthcare to finance.

The introduction of Inkling could solidify Thinking Machines' position as a significant contender in the rapidly evolving AI sector. Open-source models have gained popularity due to their cost-effectiveness compared to proprietary models, which typically require a fee for access. The open-source nature of Inkling allows smaller companies and independent researchers to leverage cutting-edge technology without the financial barriers often associated with proprietary models. Additionally, open-source models can be easily tailored for specific tasks, which is a significant advantage for organizations with unique needs or niche applications. While the leading open-weight models currently originate from China, Thinking Machines asserts that Inkling delivers performance on par with these models, potentially shifting the balance of power in the global AI landscape.

This release aligns with the vision that Thinking Machines articulated in a recent blog post, advocating for a decentralized approach to AI technology. The company believes that AI should not be monopolized by a handful of corporations, enabling more individuals to create their own models using their own data. This perspective resonates with a broader movement within the tech community, which seeks to promote transparency and accessibility in AI development. The notion that AI should be a collective resource rather than an exclusive asset has gained traction, particularly among advocates for open-source software and ethical AI practices.

A source from the company, who requested anonymity to discuss the development process, revealed an intriguing phenomenon encountered during Inkling's training. Unlike other models, which typically provide natural language explanations for their complex reasoning, Inkling opted to eliminate this feature for efficiency. "It determined that the grammar was overhead, which is interesting," the source noted. This decision highlights the ongoing trade-offs in AI design, where efficiency and performance must be balanced with the need for explainability. The company later reinstated natural language reasoning to enhance the explainability of the model's decisions, acknowledging the importance of transparency in AI systems. As AI continues to permeate various aspects of society, the ability to understand how decisions are made becomes increasingly critical, especially in applications affecting human lives.

Founded in February 2025 by prominent figures from OpenAI, including Mira Murati, who previously served as CTO (and briefly CEO), John Schulman, a cofounder instrumental in developing ChatGPT, and Lilian Weng, a former VP focused on safety and robotics, Thinking Machines has made significant strides in the AI landscape. The collective expertise of its founders brings a wealth of knowledge and experience that is likely to influence the direction of the company and its products. Their backgrounds in AI safety and ethics, combined with technical prowess, position Thinking Machines to navigate the complex challenges associated with developing powerful AI technologies responsibly.

The startup has secured the largest seed funding round in history, valuing it at $12 billion right from the start. This unprecedented financial backing underscores the confidence investors have in the company's potential to innovate and disrupt the AI market. Prior to launching Inkling, the company introduced Tinker, a tool for fine-tuning models, showcased a tool for natural voice interactions, and published research in machine learning. These initiatives reflect Thinking Machines' commitment to advancing AI technology and making it more accessible to a broader audience.

While OpenAI initiated the AI boom with ChatGPT, companies led by defectors like Thinking Machines and Anthropic are rapidly gaining ground. Anthropic recently filed for an IPO and is valued at over a trillion dollars, with its model Claude gaining popularity among businesses, particularly for its coding capabilities. This competitive landscape illustrates the dynamic nature of the AI industry, where new players can emerge rapidly and challenge established entities. The success of these companies may also indicate a shift in how AI technologies are developed and deployed, as more organizations embrace open-source principles and prioritize collaboration over competition.

As the AI sector continues to evolve, the launch of Inkling by Thinking Machines Lab marks a pivotal moment in the ongoing discourse surrounding AI development. The implications of open-source AI models extend beyond just technological advancements; they also touch upon ethical considerations, regulatory frameworks, and the future of work. As more individuals and organizations gain access to powerful AI tools, the potential for innovation increases, but so too does the responsibility to use these tools ethically and responsibly. The trajectory of AI development will likely be shaped by how well the industry addresses these challenges while fostering an environment of collaboration and inclusivity.