←back to thread

Qwen3-Omni-Flash-2025-12-01：a next-generation native multimodal large model

(qwen.ai)
314 points pretext | 5 comments | 10 Dec 25 16:13 UTC | HN request time: 0.862s | source
Show context
dvh ◴[10 Dec 25 16:45 UTC] No.46219993[source]▶
I asked: "How many resistors are used in fuzzhugger phantom octave guitar pedal?". It replied 29 resistors and provided a long list. Answer is 2 resistors: https://tagboardeffects.blogspot.com/2013/04/fuzzhugger-phan...
replies(5): >>46220026 #>>46220132 #>>46220178 #>>46222417 #>>46226884 #
esafak ◴[10 Dec 25 16:56 UTC] No.46220132[source]▶
This is just trivia. I would not use it to test computers -- or humans.
replies(2): >>46220344 #>>46221915 #
1. parineum ◴[10 Dec 25 17:09 UTC] No.46220344[source]▶
Everything is just trivia until you have a use for the answer.

OP provided a we link with the answer, aren't these models supposed to be trained on all of that data?

replies(2): >>46220437 #>>46221103 #
2. esafak ◴[10 Dec 25 17:16 UTC] No.46220437[source]▶
There is nothing useful you can do with this information. You might as well memorize the phone book.

The model has a certain capacity -- quite limited in this case -- so there is an opportunity cost in learning one thing over another. That's why it is important to train on quality data; things you can build on top of.

replies(1): >>46226858 #
3. DennisP ◴[10 Dec 25 18:03 UTC] No.46221103[source]▶
Just because it's in the training data doesn't mean the model can remember it. The parameters total 60 gigabytes, there's only so much trivia that can fit in there so it has to do lossy compression.
4. parineum ◴[11 Dec 25 02:14 UTC] No.46226858[source]▶
What if you are trying to fix one of these things and needed a list of replacement parts?
replies(1): >>46226936 #
5. esafak ◴[11 Dec 25 02:28 UTC] No.46226936{3}[source]▶
Not the right problem for this model. Any RAG-backed SLM would do; the important part is being backed by a search engine, like https://google.com/ai