FramePack-U：恒定内存压缩长视频，让 VLM 看懂小时级视频
Elephant in the Fridge: Constant-Memory Frame Packing for Long Video Understanding
免训练就能让 Qwen3.5-35B-A3B 看小时级长视频，三个基准最高涨近 13 个点，推理时直接插上就能用。
长视频会让 VLM 的视觉 token 序列随帧数线性增长，内存和延迟都扛不住，现有 token 削减和关键帧选择方法没有改变 O(n) 的扩展方式。FramePack-U 是一个免训练的帧打包框架，把任意长度的视频压成固定大小的视觉表示，内存开销恒定。它按帧与文本提示的语义相关性打分，再做多样性驱动的排序，把选中帧以不同分辨率打包进固定上下文窗口，不需要微调或改架构。在 LongVideoBench、MLVU、Video-MME 三个基准上，Qwen3.5-35B-A3B 在 128 帧预算下分别提升 +12.9%、+10.6%、+9.7%，超过官方模型和此前的 token 削减方法。
Elephant in the Fridge: Constant-Memory Frame Packing for Long Video Understanding
Hour-long videos, streaming feeds, and movie-length content are pushing vision-language models (VLMs) past their breaking point -- the standard recipe of encoding each frame into visual tokens yields sequences that grow linearly with video length, causing prohibitive memory consumption and inference latency. Existing approaches, such as token reduction and keyframe selection, shorten the token sequence but leave the O(n) scaling intact, failing to resolve the fundamental bottleneck. We propose FramePack-U, a strikingly simple, training-free frame packing framework that compresses arbitrarily long videos into a fixed size visual representation at constant memory cost. FramePack-U requires no fine-tuning or architectural modification -- it plugs directly into off-the-shelf VLMs at inference time. Inspired by FramePack from the generation literature, we propose a content-aware compression strategy for visual understanding under a fixed token budget. Specifically, FramePack-U (1) scores each frame by its semantic similarity to the text prompt and ranks frames in a diversity-driven manner, and (2) packs the selected frames at varying resolutions into a fixed-length context window. This concentrates the memory budget on frames that are both instruction-relevant and visually diverse. While FramePack-U can in theory pack arbitrarily many frames into the fixed budget, we identify an empirical instantiation that generalizes robustly across settings. Extensive experiments on LongVideoBench, MLVU, and Video-MME show that FramePack-U consistently outperforms full-resolution baselines and prior token reduction and keyframe selection methods across diverse model families, parameter scales, and frame budgets. Notably, on Qwen3.5-35B-A3B with a 128-frame budget, FramePack-U yields gains of +12.9\%, +10.6\%, and +9.7\% over the official model on the three benchmarks, respectively.
