Sa2VA 4B Image | Vision | AI Playground

fal-ai/sa2va/4b/image

Sa2VA is an MLLM capable of question answering, visual prompt understanding, and dense object segmentation at both image and video levels

Inference

Commercial use

Schema

Input

Prompt*

Image Url*

Hint: you can drag and drop file(s) here, or provide a base64 encoded data URL Accepted file types: jpg, jpeg, png, webp

Result

Idle

<p>  A white pickup truck  </p>   [SEG]  is parked on the side of  <p>  the red building  </p>   [SEG] , creating a unique and eye-catching contrast.<|im_end|>

Generated image - model inference by fal.ai

{
  "output": "<p>  A white pickup truck  </p>   [SEG]  is parked on the side of  <p>  the red building  </p>   [SEG] , creating a unique and eye-catching contrast.<|im_end|>",
  "masks": [
    {
      "file_size": 15724,
      "height": 1200,
      "file_name": "019c3c1e3c50446e9996f709d36debb4.png",
      "content_type": "image/png",
      "url": "https://v3.fal.media/files/monkey/6ITmhHQJ-69s-UxajrY5T_019c3c1e3c50446e9996f709d36debb4.png",
      "width": 1800
    },
    {
      "file_size": 14905,
      "height": 1200,
      "file_name": "0a1522ca410942c7ad6c73efa15b3549.png",
      "content_type": "image/png",
      "url": "https://v3.fal.media/files/monkey/IljtMxahoo9-7SUpx0fth_0a1522ca410942c7ad6c73efa15b3549.png",
      "width": 1800
    }
  ]
}