About Segment Anything 3 - Image
Segment Anything 3 - Image is a Large Vision Model (LVM). It excels in zero-shot image segmentation, detecting and segmenting all instances of objects based on text or visual prompts like points and boxes, offering a unified interface for concept-driven tasks over earlier geometry-only models.
Some other noteworthy features of Segment Anything 3 - Image include exhaustive instance detection (finding every matching object in an image) and a presence head that checks concept existence before localization to reduce false positives.
| Metric | Value |
|---|---|
| Parameter Count | 848 million |
| Multilingual | No (English text prompts) |
*Quantization is specific to the inference provider and the model may be offered with different quantization levels by other providers.
Ready to build with Segment Anything 3 - Image?
Try Segment Anything 3 - Image in the Workbench to prompt it, compare outputs, and iterate on prompts without writing any code. When you're ready to ship, call the same model from our API and build your own apps on top of it.
curl -sSf -X POST https://hub.oxen.ai/api/ai/images/edit \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $OXEN_API_KEY" \
-d '{
"model": "sam-3-image",
"input_image": "https://hub.oxen.ai/api/repos/ox/Oxen-AI-Assets/file/main/images/ox_zoom_out_1926_1076.png",
"prompt": "the ox horns",
"apply_mask": true,
"output_format": "png"
}'Request parameters
Every field Segment Anything 3 - Image accepts in the request body. See the API reference for response formats.
https://hub.oxen.ai/api/ai/images/edit| Field | Type | Required | Default | Description |
|---|---|---|---|---|
input_image | stringuri | Yes | — | Image to use as reference. Must be jpeg, png, gif, or webp. |
prompt | string | Yes | — | Text description of what you want to segment out of the image. |
apply_mask | boolean | No | true | Apply the mask on the image. |
output_format | string | No | "png" | Format of the output imagesjpegpng |