Xiaomi MiMo AI Models Launched With Environment friendly Reasoning, Small Dimension

Xiaomi on Tuesday launched an open-source reasoning-focused synthetic intelligence (AI) mannequin. Dubbed MiMo, the household of reasoning fashions innovate the optimisation of reasoning functionality in a comparatively smaller parameter measurement. This can also be the primary open-source reasoning mannequin by the tech big, and it competes with Chinese fashions akin to DeepSeek R1 and Alibaba’s Qwen QwQ-32B, and world reasoning fashions together with OpenAI’s o1 and Google’s Gemini 2.0 Flash Thinking. The MiMo household contains 4 completely different fashions, every with distinctive use circumstances.

Xiaomi’s MiMo Reasoning AI Model to Compete With DeepSeek R1

With the MiMo collection of AI fashions, Xiaomi researchers aimed to resolve the scale drawback in reasoning AI fashions. Reasoning fashions (at the very least ones that may be measured) have round 24 billion or extra parameters. The massive measurement is saved to attain uniform and simultaneous enhancements in each coding and mathematical capabilities of enormous language fashions, one thing thought of troublesome to attain with smaller fashions.

In comparability, MiMo options seven billion parameters, and Xiaomi claims that its efficiency matches OpenAI’s o1-mini and outperforms a number of reasoning fashions with 32 billion parameters. The researchers claimed that the bottom AI mannequin was pre-trained on 25 trillion tokens.

The researchers claimed that such effectivity was achieved by optimising information preprocessing pipelines, enhancing textual content extraction toolkits, and making use of multidimensional information filtering. Further, MiMo’s pre-training included a three-stage information combination technique.

Based on inner testing, the Xiaomi researchers declare that the MiMo-7B-Base scores 75.2 on the BIG-Bench Hard (BBH) benchmark for reasoning capabilities. The zero-shot reinforcement studying (RL)-based MiMo-7B-RL-Zero is claimed to excel in arithmetic and coding-related duties, and scores 55.4 on the AIME benchmark, outperforming o1-mini by 4.7 factors.

As MiMo is an open-source AI mannequin, it may be downloaded from Xiaomi’s itemizing on GitHub and Hugging Face. The technical paper particulars the mannequin’s structure in addition to the pre-training and post-training processes. It is a text-based mannequin and doesn’t have multimodal capabilities. Similar to most open-source releases, the main points concerning the mannequin’s dataset isn’t recognized.