MaLoW
Collection
MaLoW: Memory as LoRA Weights. Checkpoints for the paper "Memory as Weights: Internalizing Long-Term History for Streaming Videos". • 9 items • Updated
MaLoW (Qwen3.5-9B)
Checkpoint for Memory as Weights: Internalizing Long-Term History for Streaming Videos.
Paper: Memory as Weights: Internalizing Long-Term History for Streaming Videos
Repository: MaLoW.
Project page: MaLoW.
Download the base model separately: Qwen/Qwen3.5-9B. This repository contains inference modules, not a standalone base model.
Follow docs/video_evaluation.md in the MaLoW code release for OVO-Bench and StreamingBench. The JSON metadata records the inference configuration; load it through the MaLoW runtime.
Reported reference results (percent):
| OVO-Bench backward | Real-time | Forward | Average |
|---|---|---|---|
| 66.35 | 76.16 | 51.48 | 64.64 |
| StreamingBench average | Real-time | Omni | Proactive | SQA |
|---|---|---|---|---|
| 73.13 | 82.16 | 62.06 | 58.8 | 63.6 |
Weights and inference metadata are distributed under Apache-2.0. See LICENSE and NOTICE; base models and benchmark assets retain their respective upstream terms.
- Downloads last month
- 44
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support