TimeLens2-9B-SFT

TimeLens2-9B-SFT is a video temporal grounding model based on Qwen/Qwen3.5-9B. Given a video and a text query, it predicts the relevant time intervals in seconds. This is the SFT version.

Paper · Code & Usage · Collection

Benchmark Results

Benchmark mIoU (%)
Charades-STA 60.73
ActivityNet Captions 62.43
QVHighlights 71.85
VUE-TR 49.56
VUE-TR-v2 47.45
MomentSeeker 29.53
Ego4D-NLQ-v2 22.04
Average 49.08

The two VUE benchmarks use the available-video subsets. The average is the unweighted mean over the seven benchmarks.

For inference and evaluation, follow the official TimeLens2 repository and use ZhuYuhan/TimeLens2-9B-SFT as the model path.

Downloads last month
39
Safetensors
Model size
9B params
Tensor type
F32
·
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for ZhuYuhan/TimeLens2-9B-SFT

Finetuned
Qwen/Qwen3.5-9B
Finetuned
(1,433)
this model

Paper for ZhuYuhan/TimeLens2-9B-SFT