Answers a question about one image. It also captions and tags pictures dropped into a studio inbox so later messages can refer to them.
Visual Question Answering
Cross-sector
Tool
Productised
Local/private
2026
The problem
To find out what is in a picture, someone still had to open it. Images dropped into a working chat were invisible to later messages because nothing had described them.
What it does
One route sends a single image and a question to a vision model and returns an answer based on the picture itself, with no search of stored documents. Another route picks up images that have not been described, skips any that are too large, and saves a caption and labels with the file. A failed image is left for a rest period and not retried straight away.
Outcomes
A question about a picture gets a sentence back.
Chat images gain a caption before the next message.
Oversized files are skipped and do not hold up the waiting list.
The picture does not have to be added to a searchable document store.
Simplified mockup
Interactive preview with sample data. Nothing is sent to a server.
Ask about one picture
court-trainer.png 1.4 MB
Sent with the image. Nothing is looked up in a document index.
Direct path
Ready when you are
Answer from the pixels
The sole is gum brown with a darker heel.
Sentences
0
Index hits
0
Under 6 MB cap
Yes
Studio inbox
3 files waiting on the queue
Undescribed uploads
File
Size
State
court-trainer.png
1.4 MB
undescribed
raw-yard-scan.tif
18.6 MB
over 6 MB cap
label-blur.jpg
880 KB
cool-off 28 min
court-trainer.png stored on the file
Caption
Court trainer on a white ground
footwearwhiteleather
Queue after this pass
File
What happened
State
court-trainer.png
Caption and three labels written
stored
raw-yard-scan.tif
18.6 MB, left in the archive
skipped
label-blur.jpg
Earlier failure, not retried
cool-off 28 min
Captions later messages can use
Morning thread
Use the trainer I dropped this morning.
Court trainer on a white ground. Tagged footwear, white, leather.
File rows
File
Caption
Labels
State
court-trainer.png
Court trainer on a white ground
footwear, white, leather
stored
gum-sole-crop.png
Gum sole with a darker heel block
footwear, sole
stored
plant-room-door.jpg
Green door on a brick plant room
site, door
stored
raw-yard-scan.tif
Not described
–
skipped
label-blur.jpg
Not described
–
cool-off
How it works
Send a question and an image to the vision model.
Take the answer from the picture itself and not from stored documents.
Add an uploaded image to a waiting list of undescribed images.
Caption and label each image under the size limit with a locally run model.
Save the caption and labels with the file.
Leave a failed image alone for a rest period before trying again.
Pipeline
flowchart LR
A["Image"] --> B{"Question supplied?"}
B -->|Yes| C["Answer from pixels"]
B -->|No| D["Inbox queue"]
D --> E{"Under size cap?"}
E -->|Yes| F["Caption and tags"]
E -->|No| G["Skip"]
F --> H["Store on file"]
Output
Dummy court trainer on a white ground, used for the question and the inbox caption.