coppsilgold 15 hours ago

> The test works by inserting a semantically important "needle" frame at random positions in long videos, which the system must then find and analyze.

This seems to be somewhat unwise. Such an insertion would qualify as an anomaly. And if it's also trained that way, would you not train the model to find artificial frames where they don't belong?

Would it not have been better to find a set of videos where something specific (common, rare, surprising, etc) happens at some time and ask the model about that?

  • IanCal 3 hours ago

    That rather depends on exactly how this is done, although it's a useful upper bound for many tasks either way. You could say the same for images and yet due to the way some work they straight up cannot see in certain ways.

    This could describe adding a frame of nonsense into an existing video.

    It also could describe finding a semantically useful thing in an actual video, where the exact location is randomised by looking at different time crops of the video. For example, finding a book on a desk in a video that's only there in a panning shot, and you then see if it can find it in a 10s cut, 20s cut, 10 minute cut, etc, and near the start/middle/end.

    Here's the paper: https://arxiv.org/pdf/2511.21631

    > To evaluate the model’s capability in processing long-context inputs, we construct a video “Needle-ina-Haystack” evaluation on Qwen3-VL-235B-A22B-Instruct. In this task, a semantically salient “needle” frame—containing critical visual evidence—is inserted at varying temporal positions within a long video. The model is then tasked with accurately locating the target frame from the long video and answering the corresponding question. During evaluation, videos are uniformly sampled at 1 FPS, and frame resolution is dynamically adjusted to maintain a constant visual token budget.

    This potentially sounds more like the former, but I can't find more accurate information on how this works.

    Regardless I'd say again that while not the whole story things like this really are useful to know, and can be very important to test - it's really not a given that models can always find anything in their context window, perhaps even more so for video.

  • bigmadshoe 4 hours ago

    Yeah the needle in a haystack tests are so stupid. It seems clear with LLMs that performance degrades massively with context size, yet those tests claim the model performs perfectly.

    • patates 3 hours ago

      As someone who abuses gemini regularly with a 90% full context, the model performance does degrade for sure but I wouldn't call it massively.

      I can't show any evidence as I don't have such tests, but it's like coding normally vs coding after a beer or two.

      For the massive effect, fill it 95% and we're talking vodka shots. 99%? A zombie who can code. But perhaps that's not fair when you have 1M token context size.

mikae1 17 hours ago

Hope this on day will be used for auto-tagging all video assets with time codes. The dream of being able to search for running horse and find a clip containing a running horse at 4m42s in one of thousands of clips.

  • tontonius 4 hours ago

    this is a solved problem already — check out https://getjumper.io where you can do exactly this (search through 100s of hours) offline and locally.

    Disclaimer: co-founder

  • ArnavAgrawal03 13 hours ago

    you can do that with Morphik already :)

    We use an embedding model that processes videos and allows you to perform RAG on them.

    • eurekin 6 hours ago

      Would it allow me to query my library for every movie that contains dance routing move1-move2-move3 in that order?

    • bn-l 9 hours ago

      Rag as in the content is used to generate an answer or rag as in searching for a video?

  • laidoffamazon 15 hours ago

    It’s not difficult to hack this together with CLIP. I did this with about a tenth of my movie collection last week with a GTX 1080 - though it lacks temporal understanding so you have to do the scene analysis yourself

    • vhcr 11 hours ago

      I'm guessing you're not storing the CLIP for every single frame, instead of every second or so? Also, are you using the cosine similarity? How are you finding the nearest vector?

      • laidoffamazon 2 hours ago

        I split per scene using pyscenedetect and sampled from each. Distance is via cosine similarity- I fed it into qdrant

    • dynode 13 hours ago

      Would you be willing to share more details of what you did?

DrAwdeOccarim 2 hours ago

Does anyone know how this actually was done? Like, did they export every frame as a PNG and then run them each one by one through the model? Or did they somehow "load" the video into the model directly (which then internally somehow steps through each frame?)

re5i5tor 12 hours ago

For anyone using Qwen3-VL: where are you running it? I had tons of reliability problems with Qwen3-VL inference providers on OpenRouter — based on uptime graphs I wasn’t alone. But when it worked, Qwen3-VL was pack-leading good at AI Vision stuff.

  • sosodev 2 hours ago

    I’ve noticed that the open weight models have a lot of issues on OpenRouter. You get a lot of inconsistency in quality due to varying quants at least. I’ve had some seriously nonsensical responses from models that I can’t replicate at all when I switch providers. Lots that just randomly fail to handle requests too. I would recommend finding a provider that works best for your needs and pinning it.

  • lreeves 5 hours ago

    I run the larger version of it on a Threadripper with 512GB RAM and a 32GB GPU for the non-expert layers and context, using llama.cpp. Performs great, however god forbid you try to get that much memory these days.

  • m00dy 12 hours ago

    I run it on ollama

    • nicman23 11 hours ago

      the big boy model?

      • adastra22 11 hours ago

        It's not that big of a model?

        • mkl 11 hours ago

          235B-A22B is pretty big.

clusterhacks 16 hours ago

I was playing around with Qwen3-VL to parse PDFs - meaning, do some OCR data extraction from a reasonably well-formated PDF report. Failed miserably, although I was using the 30B-A3B model instead of the larger one.

I like the Qwen models and use them for other tasks successfully. It is so interesting how LLMs will do quite well in one situation and quite badly in another.

neves 5 hours ago

My favorite AI feature is to put a YouTube link in Gemini and ask it to summarize. Or even better: put the link of a 20min video "5 ways to" and ask "what are the 5 ways?"

I think Gemini analyzes the transcription.

Can I do the same for free with Qwen3?

  • seidleroni an hour ago

    I don't believe that it just analyzes the transcription. I asked Gemini to look at the youtube video referenced on the site below and "build" something that duplicates that device. It did a pretty good approximation that it could not have done without going through the full video.

    https://bitsnpieces.dev/posts/a-synth-for-my-daughter/

djmips 19 hours ago

Does anyone else worry about this technology used for Big Brother type surveillance?

  • reactordev 19 hours ago

    Where have you been the last decade? It’s already in use, or models like it, by companies selling access to The State

    https://deflock.me

    Not to mention cloud platforms that collect evidence and process it with all the models and store that information for searching…

    https://www.revir.ai

  • PunchyHamster 17 hours ago

    It was already used before current AI explosion.

    This is why keeping our governments from eating that tasty apple of "if you can record AND analyse everything there will be so much less crime" and "just give us keys to all private communication, we swear we will just use it to find bad guys". Because someone will, and someone will use it to hit on people they don't like

  • kelipso 16 hours ago

    This tech would be a massive waste of computational resources to do that. Technology for what you said is way more efficient and has been working well for years now.

  • basilgohar 18 hours ago

    How do you think this tech was developed in the first place? It's probably trained and used in the surveillance bid for a decade before it comes to consumers, and this probably isn't the SoA stuff that governments have access to, we're probably 5-10 years behind what's on the cutting edge.

    • speedgoose 9 hours ago

      I wouldn’t bet. IT innovation used to be lead by the defence industry, but that has changed and now consumer technology is driving the innovation from what I have been told.

      I’m sure they have some cool secret stuff, but they are perhaps not 10 years ahead. Also, I find unlikely that those secrets wouldn’t make it to the public society now, as we are probably close the top of the AI bubble.

  • protocolture 18 hours ago

    We got Facial Rec and LPR first, those are more dangerous for surveillance.

  • g-mork 19 hours ago

    warmly encourage you avoid reading the header files of the dahua camera SDK

    • 71bw 5 hours ago

      Mind sharing a bit more insight?

  • bgwalter 16 hours ago

    In surveillance and police states like The Netherlands it has been used since forever:

    https://www.theguardian.com/cities/2018/mar/01/smart-cities-...

    Now people will say again that this project has been abandoned, which just isn't true (2024):

    https://www.dutchnews.nl/2024/06/smart-street-surveillance-o...

    • thijson 15 hours ago

      I was watching a crime solving show from the UK. A huge percentage of the crimes are solved using camera footage. Also, they use geofencing, looking at which phones went in and out of the crime location at the time of the crime.

      • fy20 13 hours ago

        I would be surprised if this hasn't existed for a few decades already.

        Back in 2009 I was working at a place where O2 was a client, and they gave us an API that could identify the cell tower (inc. lat/lng) any of their customers were connected to. The network needs to track this data internally to function, so the API is basically the equivalent of their DNS.

  • ants_everywhere 17 hours ago

    Big Brother is a reference to George Orwell's critique of Communism in Nineteen Eighty-Four.

    Qwen is a video model trained by a Communist government, or technically by a company with very close ties to the Chinese government. The Chinese government also has laws requiring AI be used to further the political goals of China in particular and authoritarian socialism in general.

    In the light of all this, I think it's reasonable to conclude that this technology will be used for Big Brother type surveillance and quite possible that it was created explicitly for that purpose.

    • Intermernet 15 hours ago

      Just nitpicking here, but 1984 is a critique of totalitarianism. The only references to systems of government in the book refer to "The German Nazis and the Russian Communists".

      Orwell was a democratic socialist. He was opposed to totalitarian politics, not communism per se.

      • ants_everywhere 14 hours ago

        It's true that it's about totalitarianism to some extent. But we have Orwell's actual words here that it's chiefly about communism

        > [Nineteen Eighty-Four] was based chiefly on communism, because that is the dominant form of totalitarianism, but I was trying chiefly to imagine what communism would be like if it were firmly rooted in the English speaking countries, and was no longer a mere extension of the Russian Foreign Office.

        And of course Animal Farm is only about communism (as opposed to communism + fascism). And the lesser known Homage to Catalonia depicts the communist suppression of other socialist groups.

        By all this I just mean to say when you're reading Nineteen Eighty-Four what he's describing is barely a fictionalization of what was already going on in the Soviet Union. There's just not a lot in the book that is specifically Nazi or Fascist.

        I don't have any opinion on whether he thought there were non-totalitarian forms of communism.

        • justsomejew 2 hours ago

          I think that Orwell understood his own people much more than Russians, so it might be useful, while reading him, to take a look at the mirror as well..

Alifatisk 6 hours ago

Many of the models we have today seem to only perform OCR on the images you send and use the text retrieved for context when answering. However, Qwen-VL, and I guess Gemini now? Are different, they seem to "understanding" the image I send with my prompt. They manage to capture spatial relationships, objects, and semantics from the image, it's very impressive. I’ve been telling my friends about the Qwen3-VL model option in Qwen Chat for a while because I feel like it’s underrated.

chhxdjsj 14 hours ago

Not so relevant to the thread but ive been uploading screenshots from citrix guis and asking qwen3-vl for the appropriate next action eg Mouseclick, and while it knows what to click it struggles to accurately return which pixel coordinates to click. Anyone know a way to get accurate pixel coordinates returned?

  • spherelot 11 hours ago

    How do you prompt the model? In my experience, Qwen3-VL models have very accurate grounding capabilities (I’ve tested Qwen3-VL-30B-A3B-Instruct, Qwen3-VL-30B-A3B-Thinking, and Qwen3-VL-235B-A22B-Thinking-FP8).

    Note that the returned values are not direct pixel coordinates. Instead, they are normalized to a 0–1000 range. For example, if you ask for a bounding box, the model might output:

    ```json [ {"bbox_2d": [217, 112, 920, 956], "label": "cat"} ] ```

    Here, the values represent [x_min, y_min, x_max, y_max]. To convert these to pixel coordinates, use:

    [x_min / 1000 * image_width, y_min / 1000 * image_height, x_max / 1000 * image_width, y_max / 1000 * image_height]

    Also, if you’re running the model with vLLM > 0.11.0, you might be hitting this bug: https://github.com/vllm-project/vllm/issues/29595

    • chhxdjsj 11 hours ago

      Will give this a go, cheers :)

  • jazzyjackson 14 hours ago

    Could you combine it with a classic OCR segmentation process, so that along with the image you also provide box coordinates of each string?

  • hamasho 13 hours ago

    It's very not accurate, but sometimes instructing to return pyautogui code works.

      prompt: I attach a screenshot (1920x1080). Write code to click the submit button using pyautogui.
      attachment: <screenshot>
      reply:
        import pyautogui
        pyautogui.click(100, 200)
    • chhxdjsj 11 hours ago

      Ive been asking for pyautogui output already but it is still very hit and miss

  • visioninmyblood 13 hours ago

    you want get the exact coordinated by running a key point network to pinpoint which coordinates does the next click point is you can. here I show a example simple prompt which returns the keypoint location of the next botton to click and visually localize the point with a keypoint in the image

    https://chat.vlm.run/c/e12f0153-7121-4599-9eb9-cd8c60bbbd69

ta12653421 3 hours ago

so, a great politician once said: "the internet is for p0rn" - and if we accept this, what will this thing pinpoint in p0rn videos? :-D LOL

  • spider-mario 32 minutes ago

    There it is! Oh, it disappeared. There it is again! Oh, it disappeared. It’s back! Wait, no.

visioninmyblood 19 hours ago

I was using this for video understanding with inference form vlm.run infra. It definitely has outperformed Gemini which generally is much better than openai or Claude on videos. The detailed extraction is pretty good. With agents you can also crop into a segment and do more operations on it. have to see how the multi modal space progresses:

link to results: https://chat.vlm.run/c/82a33ebb-65f9-40f3-9691-bc674ef28b52

Quick demo: https://www.youtube.com/watch?v=78ErDBuqBEo

  • colechristensen 17 hours ago

    I found it pretty funny how bad Claude was at cropping an image. It was a cute little character with some text off to the side on a white background, all very clean cartoon vibes and it COULD NOT just select the character. I pursued it for 20 minutes because I thought it was funny. Of course it was 45 seconds to do it myself.

    A lot of my side projects involve UIs and almost all of my problems with getting LLMs to write them for me involve "The UI isn't doing what you say it's doing" and struggling to get A) a reliable way to get it to look at the UI so it can continue its loop and B) getting it to understand what it's looking at well enough to do something about it

    • visioninmyblood 17 hours ago

      I agree claude and chatgpt and even gemini does a poor job in detecting and cropping into a region. Some of the simplest tasks, Qwen also is great at summerization but not into solving simple vision tasks like cropping, segmentetation and detection. Here is an examples where we compared claude, gemini, chatgpt and other frontier models for simple(and complicated) visual tasks https://chat.vlm.run/showdown#:~:text=Crop%20into%20the%20cl...

      • colechristensen 17 hours ago

        The part that was funny to me is I would respond "is that right?" and it would tell me exactly how it was wrong and proceed to do it incorrectly again in a very similar but different way. It was like a Monty Python sketch. I might have also been very tired and easily amused.

eurekin 19 hours ago

Insane if true... now I wonder, if I use it to go through some old dance routing video catalogue to recognize and write individual move lists

thot_experiment 3 days ago

anyone have a tl;dr for me on what the best way to get the video comprehension stuff going is? i use qwen-30b-vl all the time locally as my goto model because it's just so insanely fast, curious to mess with the video stuff, the vision comprehension works great and i use it for OCR and classification all the time

  • xrd 19 hours ago

    How much VRAM do you need for local usage may I ask?

m00dy 12 hours ago

Ive used qwen3-VL on deepwalker lately. All I can stay is that this model is so underrated.

[0]: https://deepwalker.xyz

CSMastermind 13 hours ago

Still not great at the use cases I tested it for but Gemini isn't either. I think we're still very early on video comprehension.

spwa4 19 hours ago

It's so weird how that works with transformers.

Finetuning an LLM "backbone" (if I understand correctly: a fully trained but not instruction tuned LLM, usually small because students) with OCR tokens bests just about every OCR network out there.

And it's not just OCR. Describing images. Bounding boxes. Audio, both ASR and TTS, all works better that way. Now many research papers are only really about how to encode image/audio/video to feed it into a Llama or Qwen model.

  • zmmmmm 19 hours ago

    It is fascinating. Vision language models are unreasonably good compared to dedicated OCR and even the language tasks to some extent.

    My take is it fits into the general concept that generalist models have significant advantages because so much more latent structure maps across domains than we expect. People still talk about fine tuning dedicated models being effective but my personal experience is it's still always better to use a larger generalist model than a smaller fine tuned one.

    • kgeist 17 hours ago

      >People still talk about fine tuning dedicated models being effective

      >it's still always better to use a larger generalist model than a smaller fine tuned one

      Smaller fine-tuned models are still a good fit if they need to run on-premises cheaply and are already good enough. Isn't it their main use case?

      • bangaladore 17 hours ago

        Latency and size. Otherwise pretty much useless.

    • jepj57 17 hours ago

      Now apply that thinking to human-based neural nets...

moralestapia 20 hours ago

To me, this qualifies as some sort ASI already.