Ovis2.5 技术报告

摘要

我们推出Ovis2.5，作为Ovis2的继任者，专为原生分辨率视觉感知与强大多模态推理而设计。Ovis2.5集成了一个原生分辨率视觉Transformer，能够以图像的原生可变分辨率进行处理，避免了固定分辨率分块带来的质量下降，同时保留了精细细节与全局布局——这对于复杂图表等视觉密集内容至关重要。为增强推理能力，我们训练模型超越线性思维链，执行反思——包括自我检查与修正。这一高级能力在推理时作为可选的“思考模式”呈现，允许用户在延迟与困难输入上的准确性之间做出权衡。模型通过一个全面的五阶段课程进行训练，逐步构建其技能。该过程始于基础视觉与多模态预训练，经过大规模指令调优，最终利用DPO和GRPO进行对齐与推理增强。为高效扩展这些升级，我们采用多模态数据打包与混合并行策略，实现了显著的端到端加速。我们发布了两款开源模型：Ovis2.5-9B与Ovis2.5-2B。后者延续了Ovis2“小模型，大性能”的理念，非常适合资源受限的端侧场景。在OpenCompass多模态排行榜上，Ovis2.5-9B平均得分78.3，较其前身Ovis2-8B有显著提升，并在40B参数以下的开源MLLM中达到顶尖水平；Ovis2.5-2B得分73.9，确立了其规模下的SOTA地位。除了综合得分，Ovis2.5在STEM基准测试中取得领先成绩，在基础任务与视频任务上展现出强大能力，并在复杂图表分析方面实现了其规模下的开源SOTA。

English

We present Ovis2.5, a successor to Ovis2 designed for native-resolution visual perception and strong multimodal reasoning. Ovis2.5 integrates a native-resolution vision transformer that processes images at their native, variable resolutions, avoiding the degradation from fixed-resolution tiling and preserving both fine detail and global layout -- crucial for visually dense content like complex charts. To strengthen reasoning, we train the model to move beyond linear chain-of-thought and perform reflection -- including self-checking and revision. This advanced capability is exposed as an optional "thinking mode" at inference time, allowing users to trade latency for enhanced accuracy on difficult inputs. The model is trained via a comprehensive five-phase curriculum that progressively builds its skills. The process begins with foundational visual and multimodal pretraining, advances through large-scale instruction tuning, and culminates in alignment and reasoning enhancement using DPO and GRPO. To scale these upgrades efficiently, we employ multimodal data packing and hybrid parallelism, yielding a significant end-to-end speedup. We release two open-source models: Ovis2.5-9B and Ovis2.5-2B. The latter continues the "small model, big performance" philosophy of Ovis2, making it ideal for resource-constrained, on-device scenarios. On the OpenCompass multimodal leaderboard, Ovis2.5-9B averages 78.3, marking a substantial improvement over its predecessor, Ovis2-8B, and achieving state-of-the-art results among open-source MLLMs in the sub-40B parameter range; Ovis2.5-2B scores 73.9, establishing SOTA for its size. Beyond aggregate scores, Ovis2.5 achieves leading results on STEM benchmarks, exhibits strong capabilities on grounding and video tasks, and achieves open-source SOTA at its scale for complex chart analysis.

Ovis2.5 技术报告

Ovis2.5 Technical Report

摘要

Support