According to ZDXX on June 24th, the Xiaomi MiMo Open Platform announced an extension of the chat experience and API access experience period for its MiMo-V2.5-Pro-UltraSpeed model, which was launched on June 9th. The original experience window was scheduled to end on June 23rd, but due to the number of applications far exceeding expectations, the team decided to extend the opening time. Official data shows that as of June 23rd, the model has received over 66,000 usage applications. Applicants include Fortune Global 500 companies, industry-leading enterprises, and individual developers, covering fields such as law, finance, communications, logistics, automobile manufacturing, culture and media, and universities. The Xiaomi MiMo team stated the number of applications was "far beyond expectations." The MiMo-V2.5-Pro-UltraSpeed is an ultra-fast inference mode jointly launched by the Xiaomi MiMo team and the AI inference system team TileRT, achieving an output speed of over 1000 tokens/s on a flagship model with one trillion parameters (1T), with a peak of about 1200 tokens/s. The model is based on the MoE architecture, with a total of 1T parameters and activated parameters of about 42 billion in a single forward propagation, supporting a super-long context of 1 million tokens.