Qwen2-VL 72B Instruct is a multimodal LLM designed for complex visual and textual understanding, excelling in image, video, and text-based reasoning tasks. It excels in handling diverse image resolutions, analyzing videos up to 20 minutes in length, and supporting multilingual input—including English, Chinese, and most European, Asian, and Middle Eastern languages.

Some other noteworthy features of Qwen2-VL 72B Instruct include its ability to generate descriptive text from images and videos, perform multimodal question answering, and operate devices such as mobile phones and robots based on combined visual and textual instructions.

Metric	Value
Parameter Count	72 billion
Mixture of Experts	No
Context Length	128,000 tokens
Multilingual	Yes
Quantized*	No

*Quantization is specific to the inference provider and the model may be offered with different quantization levels by other providers.