TimeLens2-8B
TimeLens2-8B is a video multimodal large language model for temporal grounding. Given a video and a text query, it returns the time interval containing the relevant visual evidence.
The model is built on Qwen3-VL-8B-Instruct and achieves 48.0 average mIoU across seven temporal grounding benchmarks.
TimeLens2-8B sets a new state of the art on this seven-benchmark suite, with strong performance across short-, long-, and egocentric-video grounding.
Paper
TimeLens2: Generalist Video Temporal Grounding with Multimodal LLMs
Benchmark Results
Inference
pip install -U torch torchvision "transformers>=4.57.0" accelerate "qwen-vl-utils[decord]>=0.0.14"
pip install -U flash-attn --no-build-isolation
from pathlib import Path
from qwen_vl_utils import process_vision_info
from transformers import AutoModelForImageTextToText, AutoProcessor
model_id = "MCG-NJU/TimeLens2-8B"
video_path = "/path/to/video.mp4"
query = "A man opens the refrigerator."
model = AutoModelForImageTextToText.from_pretrained(
model_id,
torch_dtype="auto",
device_map="auto",
attn_implementation="flash_attention_2",
)
processor = AutoProcessor.from_pretrained(model_id)
prompt = (
f'Given the query: "{query}", return ALL time spans (in seconds) where the query is relevant.\n'
"Output format MUST be a JSON array of [start, end] pairs.\n"
)
messages = [
{
"role": "user",
"content": [
{
"type": "video",
"video": Path(video_path).resolve().as_uri(),
"fps": 2.0,
"min_pixels": 32 * 32,
"max_pixels": 480 * 480,
"total_pixels": 128000 * 32 * 32,
},
{"type": "text", "text": prompt},
],
}
]
text = processor.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True,
)
images, videos, video_kwargs = process_vision_info(
messages,
image_patch_size=16,
return_video_kwargs=True,
return_video_metadata=True,
)
if videos is not None:
videos, video_metadatas = zip(*videos)
videos, video_metadatas = list(videos), list(video_metadatas)
else:
video_metadatas = None
inputs = processor(
text=text,
images=images,
videos=videos,
video_metadata=video_metadatas,
do_resize=False,
return_tensors="pt",
**video_kwargs,
).to(model.device)
output_ids = model.generate(
**inputs,
max_new_tokens=4096,
temperature=0.01,
top_p=0.001,
top_k=1,
repetition_penalty=1.0,
)
output_ids = [
output[len(input_ids) :]
for input_ids, output in zip(inputs.input_ids, output_ids)
]
response = processor.batch_decode(
output_ids,
skip_special_tokens=True,
clean_up_tokenization_spaces=False,
)
print(response[0])
Citation
@article{zhu2026timelens2,
title={TimeLens2: Generalist Video Temporal Grounding with Multimodal LLMs},
author={Zhu, Yuhan and Ma, Changlian and Zeng, Xiangyu and Li, Xinhao and Zhang, Zhiqiu and Li, Songze and Zhang, Jun and Jiang, Tianxiang and Yang, Yuandong and Yan, Ziang and others},
journal={arXiv preprint arXiv:2607.17423},
year={2026}
}