Inception, a new Palo Alto-based company founded by Stanford computer science professor Stefano Ermon, has announced a novel AI model based on "diffusion" technology, which promises to revolutionize the field of artificial intelligence. The company's diffusion-based large language model, or DLM, offers the capabilities of traditional large language models (LLMs) but with significantly faster performance and reduced computing costs.

The development of Inception's DLM is a significant breakthrough in the field of AI, as traditional LLMs have been limited by their sequential processing approach. According to Ermon, traditional LLMs work by generating one word at a time, whereas diffusion models can generate large blocks of text in parallel. This parallel processing approach enables Inception's DLM to achieve faster performance and reduced computing costs compared to traditional LLMs.

Ermon's research on diffusion models began in his Stanford lab, where he explored the possibility of applying diffusion technology to text generation. After years of research, Ermon and one of his students achieved a major breakthrough, which they detailed in a research paper published last year. Recognizing the potential of their discovery, Ermon founded Inception last summer, tapping two former students, UCLA professor Aditya Grover and Cornell professor Volodymyr Kuleshov, to co-lead the company.

Inception has already secured several customers, including unnamed Fortune 100 companies, by addressing their critical need for reduced AI latency and increased speed. The company offers an API as well as on-premises and edge device deployment options, support for model fine-tuning, and a suite of out-of-the-box DLMs for various use cases. According to Inception, its DLMs can run up to 10x faster than traditional LLMs while costing 10x less.

Inception's claims are impressive, with its "small" coding model reportedly as good as OpenAI's GPT-4o mini while being more than 10 times as fast. The company's "mini" model is said to outperform small open-source models like Meta's Llama 3.1 8B and achieves more than 1,000 tokens per second. If Inception's claims hold up, its DLM could have a significant impact on the field of AI, enabling faster and more efficient text generation.

The development of Inception's DLM is a significant step forward in the field of AI, and its potential applications are vast. With its faster performance and reduced computing costs, Inception's DLM could enable new use cases for AI, from chatbots and virtual assistants to language translation and content generation. As the field of AI continues to evolve, Inception's DLM is likely to play a significant role in shaping its future.

While Inception has made significant progress in developing its DLM, the company's funding remains unclear. According to TechCrunch, the Mayfield Fund has invested in Inception, but the exact amount of funding remains undisclosed. Despite this, Inception's DLM is an exciting development in the field of AI, and its potential impact could be significant.