Moore Threads co-founder Wang Dong: There’s no “universal chip” in the reasoning market—only a combination of solutions

Golden Finance reported that on July 18, Wang Dong, co-founder and executive chairman of GPU manufacturer Moore Threads, said, “The development of large models is very rapid both domestically and internationally. Now, top vendors complete an iteration of cutting-edge foundation model versions every two months on average. As for the cost of model deployment, we have found that China’s cutting-edge foundation models have a clear cost advantage over overseas models at the same level of intelligence, and China’s models offer better value for money. This also shows that, given limited compute resources, model companies have done a great deal of work on how to improve model efficiency, price efficiency, and training costs.”

Wang Dong noted that there is no “universal chip” in the inference market, but rather a combination of “solutions.” “The technical application threshold for the inference market is relatively low, and the scenarios are highly fragmented, so no single company can monopolize all sub-application scenarios. There is no absolutely perfect single hardware. Through flexible co-optimization between software and hardware, each model can find the hardware combination that best suits it, achieving the best balance between cost and performance. The market will see a large number of ISP companies emerge, providing MaaS providers or end customers with more cost-effective and more flexible customized inference services.”