smart_toy
florence-2-large
wowmachine/florence-2-large
Florence-2 Large is Microsoft's specialist image-captioning model — pass one image, get a text description back. Pick the caption verbosity (`Caption` for alt-text, `Detailed` for a sentence, `More Detailed` for a paragraph). Fast, cheap, ideal for batch metadata work on product catalogues.
Input
Image to caption. Florence-2 reads the picture and returns a text description.
add_photo_alternate
Caption verbosity. `Caption` returns a short alt-text-style line; `Detailed Caption` returns 1-2 sentences; `More Detailed Caption` returns a full paragraph.
Output
smart_toy
Run the model to see output here.
Pricing
**1 credit per call.** Flat per-call cost across the three verbosity tiers.
*Each call processes one image and returns one caption. Run the model multiple times for batch catalogue coverage.*...
Features
## What Florence-2 Large is good at
Florence-2 Large is Microsoft's open-source vision model purpose-built for one job:
**captioning images**. You pass one image and select the verbosity tier; the mo...
API Examples
curl -X POST \
/api/v1/model/wowmachine/florence-2-large \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"input": {}
}'