We ran seven tests across five state-of-the-art Large Multimodal Models (LMMs) on November 23rd, 2023. QwenVL passed at five of seven tests and LLaVA passed at one of seven tests. Here are the results:
.png)
Based on our tests, QwenVL performs better across different multimodal tasks than LLaVA.
Provide your own image below to test YOLOv8 and YOLOv9 model checkpoints trained on the Microsoft COCO dataset.
COCO can detect 80 common objects, including cats, cell phones, and cars.
