en.Wedoany.com Reported - Xiaomi MiMo, in collaboration with TileRT, has officially launched the UltraSpeed mode for Xiaomi MiMo-V2.5-Pro. Through extreme co-design of the model and system, it achieves a generation speed exceeding 1000 tokens/s for trillion-parameter models on general-purpose GPUs for the first time.

The MiMo-V2.5-Pro-UltraSpeed API is now available simultaneously, offered at a limited-time trial price. It is priced at 3x the cost of MiMo-V2.5-Pro, while providing approximately 10x faster output speed (API experience only, not supported by Token Plan). For reference, Xiaomi MiMo-V2.5-Pro costs 0.025 RMB per million tokens for input (cache hit) / 3 RMB (cache miss), and 6 RMB per million tokens for output.

Due to limited high-speed inference resource supply, MiMo-V2.5-Pro-UltraSpeed is available on an application-based, time-limited basis. Approved users can access the API experience for a limited time, valid only from June 9, 2026, to June 23, 2026, at 23:59. Given resource constraints, trial slots are limited, and submission of an application does not guarantee review timeliness or approval rate. Xiaomi will prioritize reviewing enterprises and professional developers with genuine business needs. Approved users will receive a limited-time free Chat experience (valid within a two-week access window). To ensure experience quality and usage fairness under resource-limited conditions, the trial service rules are as follows: each account can successfully enter the queue up to 10 times per day; the maximum session duration is 30 minutes; if a session remains idle for more than 5 minutes, the system will automatically release resources.

This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com