Jev 1.14 from @typesafeai can't do this.
TypeLLM can detect every vehicle in an image and return an array of bounding boxes — including @Tesla's new flying Roadster.
Try it now: typellm.ai/cookbook/locat…
TypeLLM can detect every vehicle in an image and return an array of bounding boxes — including @Tesla's new flying Roadster.
Try it now: typellm.ai/cookbook/locat…
Oct 16, 2026 · 5:44 AM UTC
8
11
232
27,375
@typesafeait@Teslaev only supports one bounding box per question.
This means that if an image contains multiple objects of the same category, Jev cannot return all their bounding boxes in a single call, since the questions are independent.
You can test this by uploading an image containing several dogs and asking Jev to return all their bounding boxes.
docs.typesafe.ai/primitives/box
This means that if an image contains multiple objects of the same category, Jev cannot return all their bounding boxes in a single call, since the questions are independent.
You can test this by uploading an image containing several dogs and asking Jev to return all their bounding boxes.
docs.typesafe.ai/primitives/box
1
0
4
737
Snap a photo of a receipt. WeFinance pulls out the date, merchant, category and amount, then answers things like where did my money go this month. It never asks for a bank login. Free in Anna: anna.partners/apps
0
0
2
6,353
@TypeLLM @typesafeai No doubt about your platform's capabilities, but claiming that using Jev or a Jev-like model within a harness makes you superior to Jev is unwarranted. This assertion is completely inappropriate, as you are comparing a model with a harness.
1
0
0
53
@iam_aaryanverma @typesafeai This is not done by harness; it is the native capability of the model inference, and it does not use a Jev-like model, and we are open-source: github.com/TypeLLM/TypeLLM
1
0
1
39
@Ranicket @typesafeai @Tesla No, it can't; it only supports one box per question docs.typesafe.ai/primitives/box. There is no way it can count and generate all the boxes of the same category (e.g. cars). You can try yourself by uploading an image with two or more cats and let it generate all the boxes.
1
0
1
302
@TypeLLM @typesafeai @Tesla Can we use this for shop lifting scenario in a store through cctv footage? Also what it will cost?
0
0
0
24