DiffusionGemma-as-Jev (aka djev) running near-real-time vision detection from a mobile phone using…
@mmastrac12K followers
105,036 views
15 Comments
Comments from the original post on X (Twitter)

@mmastrac
You can even try it yourself! This is the shipping docker container with a packaged server and examples:
29 likes

@mmastrac
Each image is captured from the webcam, tokenized by the model, and the structured query gets answered by DiffusionGemma in about ~300-400ms. This is the stock DiffusionGemma model, no fine tuning, running practically stock vLLM with only a few patches.
15 likes

@mpetch
It didn't do very well identifying the owner of the house (the furry one). Pet????? I'm sure th cat has something to say about that ;-)
2 likes

@ininchch
it could be useful for blind
2 likes

@Richelle_Ji
We just shipped open weights jev:
2 likes

@mmastrac
We should chat - would be great to share a single server implementation and make sure we align on multimodal Jev (mmJev?)! Feel free to reach out (DM or email)

@Richelle_Ji
dm me!

@jbermudez5
Was setting up a pipeline with YOLO26 + Jev but going to run this and compare.
1 like

@chalkers
I’m going to put it in to my robot dog
1 like

@generativist
keep going, Matt!
1 like

@unseversever
Please give it perception of depth and try this again I'll wait

@RepresenterTh
I have been following your posts since yesterday. Amazing

@Ad3DigitalMKTIA
Very cool but to big for my 12gb vram Ç_Ç

@Simplicity
Do you want T-800s? Because that's how you get T-800s.

@user_saila
I really need to know how much vram jev uses