ComfyUI Node
GLM Vision BBox Extractor
Calls a configurable GLM vision chat-completions API for one image and returns clean desc/class/bbox JSON while preserving bbox values and coordinate order.
GLM Vision BBox Extractor
- image
- bbox_json
◄prompt输出图中所有banner,logo,品牌,平台,活动,质保,店铺相关描述的区域bbox,输出为json list,格式如下,不要输出任何其他内容/格式:
[
{
"desc": "官方旗舰店",
"class": "店铺",
"bbox": [x1,y1,x2,y2]
}
]►
◄endpointhttps://open.bigmodel.cn/api/paas/v4/chat/completions►
◄modelglm-4.6v-flash►
◄api_key►
Categorydsocr_bbox/GLM Vision BBox
Inputs (5)
| Name | Type | Default | Description |
|---|---|---|---|
| image | IMAGE | — | |
| prompt | STRING | 输出图中所有banner,logo,品牌,平台,活动,质保,店铺相关描述的区域bbox,输出为json list,格式如下,不要输出任何其他内容/格式: [ { "desc": "官方旗舰店", "class": "店铺", "bbox": [x1,y1,x2,y2] } ] | — |
| endpoint | STRING | https://open.bigmodel.cn/api/paas/v4/chat/completions | — |
| model | STRING | glm-4.6v-flash | — |
| api_key | STRING | — |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| bbox_json | STRING | — |