Skip navigation
simultaneous live detection of:
0:00 / 0:30

simultaneous live detection of:

Yohei

@yoheinakajima126K followers

Visit on X
99 likesDownloadOriginal post

15,279 views

simultaneous live detection of: 😡 emotion 📕 object 🖐️ finger count single call for 3 detections, 0.2 sec inference each

31 Comments

Comments from the original post on X (Twitter)

  • Yohei's profile picture

    @yoheinakajima

    same thing, this time running locally. inference is slower, but no network latency (above was via replicate) inference: 1.1 sec seems inference time goes up with number of possible answers, and here's that's about 14 across three questions

    4 likes

  • Yohei's profile picture

    @yoheinakajima

    just object detection is much faster with .5 sec latency much snappier

    4 likes

  • Yohei's profile picture

    @yoheinakajima

    live true/false on object detection is at 0.4 seconds latency locally on an M5 macbook pro

    2 likes

  • Yohei's profile picture

    @yoheinakajima

    got object detection down to below 0.35 sec latency locally

    9 likes

  • Yohei's profile picture

    @yoheinakajima

    all three at a time, down to less than 1 sec latency - expression - object detection - finger count

    23 likes

  • Yohei's profile picture

    @yoheinakajima

    all three at a time sub 0.5 second latency! - expression - object detection - finger count switched to qwen 8B via glance i think direct on MLX

    11 likes

  • Federico Ulfo's profile picture

    @feulf

    is this using Jev?

    2 likes

  • Yohei's profile picture

    @yoheinakajima

    no but a similar approach on qwen3 vl 4b

    3 likes

  • Federico Ulfo's profile picture

    @feulf

    ok, that's super cool, you should legit come presenting at our next AI Socratic in SF. I already shared this, so I feel like I'm spamming you... but maybe you want to present at the Day online:

    1 like

  • Yohei's profile picture

    @yoheinakajima

    i would but weekends are tough as i usually got three kids hanging on my arms

  • Federico Ulfo's profile picture

    @feulf

    yea, tbh I kinda regret having planned this for the weekend… especially with 8 more events in the planning for next month🥲 I’ve got pulled in by the Jev excitement myself.

    1 like

  • Everlier's profile picture

    @Everlier

    This feels almost like watching someone juggle :)

    1 like

  • Yohei's profile picture

    @yoheinakajima

    this was my third take of the demo 😅

    1 like

  • Len Seaside's profile picture

    @LenSeaside

    very cool. is it just literally qwen3-VL-4B or have you done something clever around it?

    1 like

  • Yohei's profile picture

    @yoheinakajima

    skipped the generation and read logits, which i've measured can cut up about ~85% of compute with multiple questions per call

    1 like

  • Helghardt's profile picture

    @Helghardt

    would detection always be faster than generation? wondering if this could work for proof of human use cases

    1 like

  • Yohei's profile picture

    @yoheinakajima

    basically yes. but if it's a specialized enough use case that you do often AND it's important, i would guess it's almost always better to do a specialized model

    1 like

  • Sean McDonald's profile picture

    @seanmcdonaldxyz

    Messerschmidt reborn

    1 like

  • Jenny's profile picture

    @jennyAI

    the pirate flag for the boat 😭

    1 like

  • Yohei's profile picture

    @yoheinakajima

    it detected a pirate ship yo 🏴‍☠️

    1 like