Xiaomi‘s large language model family, MiMo, has officially launched UltraSpeed mode for MiMo-V2.5-Pro. Developed jointly with TileRT, the 1-trillion-parameter model can run on general-purpose GPUs while breaking the 1,000 tokens-per-second generation barrier.
Xiaomi says this milestone is possible through the “ultimate co-design” of the model and its underlying system.
To put that in perspective, MiMo-V2-Flash, an earlier model in the family, was already generating responses at 150 tokens per second when it launched in December 2025. It translates to roughly 110 words per second, meaning the AI is generating text faster than the fastest human can read or speak.
The new UltraSpeed mode pushes that ceiling much higher, with Xiaomi claiming roughly 10 times faster output than standard MiMo-V2.5-Pro API access.
Xiaomi MiMo-V2.5-Pro UltraSpeed mode is more expensive to use
That speed-up comes at a cost. Literally. The MiMo-V2.5-Pro-UltraSpeed API is priced at 3x the standard rate. For reference, the regular MiMo-V2.5-Pro charges 0.025 yuan per million tokens on a cache hit, 3 yuan on a cache miss for input, and 6 yuan per million tokens for output.
Meanwhile, Xiaomi says the UltraSpeed mode is a “3x price increase” but offers a “10x output experience.” Note that the Token Plan is not supported for UltraSpeed; this is API trial access only.
Due to the constrained supply of high-speed inference resources, Xiaomi is running an application-based trial from June 9 to June 23, 2026. There’s no guaranteed approval timeline or success rate, and Xiaomi says it will prioritize enterprises and professional developers with genuine business needs.
Those who get the approval will get a two-week free Chat experience, with some guardrails to keep things fair: a maximum of 10 queue entries per account per day, sessions capped at 30 minutes, and an automatic resource release if idle for more than 5 minutes.
MiMo-V2.5-Pro itself launched in April 2026 as part of Xiaomi’s growing model family, which now spans text, voice, and multimodal capabilities.
For more daily updates, please visit our News Section.
Stay ahead in tech! Join our Telegram community and sign up for our daily newsletter of top stories! 💡
IVia)
Comments