Authors: DocVQA Organizers

Affiliation: CVC-UAB, IIIT Hyderabad

Email: docvqa@cvc.uab.es

Description: Human performance as reported in InfographicVQA paper

Authors: Seed-VL

Affiliation: ByteDance

Description: Seed-VL-1.5

Authors: Xiaomi LLM-Core

Affiliation: Xiaomi

Description: We open-source MiMo-VL-7B-SFT and MiMo-VL-7B-RL, two powerful vision-language models
delivering state-of-the-art performance in both general visual understanding and multimodal
reasoning. MiMo-VL-7B-RL outperforms Qwen2.5-VL-7B on 35 out of 40 evaluated tasks, and
scores 59.4 on OlympiadBench, surpassing models with up to 78B parameters. For GUI grounding
applications, it sets a new standard with 56.1 on OSWorld-G, even outperforming specialized
models such as UI-TARS. Our training combines four-stage pre-training (2.4 trillion tokens) with
Mixed On-policy Reinforcement Learning (MORL) integrating diverse reward signals. We identify
the importance of incorporating high-quality reasoning data with long Chain-of-Thought into
pre-training stages, and the benefits of mixed RL despite challenges in simultaneous multi-domain
optimization. We also contribute a comprehensive evaluation suite covering 50+ tasks to promote
reproducibility and advance the field. The model checkpoints and full evaluation suite are
available at https://github.com/XiaomiMiMo/MiMo-VL.