Most active commenters
  • declan_roberts(3)

←back to thread

Qwen2.5-VL-32B: Smarter and Lighter

(qwenlm.github.io)
544 points tosh | 28 comments | 24 Mar 25 18:35 UTC | HN request time: 1.071s | source | bottom
Show context
simonw ◴[24 Mar 25 18:52 UTC] No.43464227[source]▶
Big day for open source Chinese model releases - DeepSeek-v3-0324 came out today too, an updated version of DeepSeek v3 now under an MIT license (previously it was a custom DeepSeek license). https://simonwillison.net/2025/Mar/24/deepseek/
replies(5): >>43464375 #>>43464498 #>>43464686 #>>43465383 #>>43467111 #
1. jschoe ◴[25 Mar 25 01:07 UTC] No.43467111[source]▶
And it still can't answer this:

Q: "9.11 and 9.9, which one is larger?"

A: "To determine which number is larger between 9.11 and 9.9, let's compare them step by step. Both numbers have the same whole number part: 9. 9.11 has a 1 in the tenths place. 9.9 has a 9 in the tenths place. Since 9 (from 9.9) is greater than 1 (from 9.11), we can conclude that 9.9 is larger than 9.11."

"Final Answer: 9.9"

I don't think anything else needs to be said here.

replies(17): >>43467119 #>>43467120 #>>43467131 #>>43467132 #>>43467138 #>>43467147 #>>43467148 #>>43467381 #>>43467419 #>>43467451 #>>43467459 #>>43467485 #>>43467500 #>>43467530 #>>43467732 #>>43467781 #>>43468916 #
2. MiiMe19 ◴[25 Mar 25 01:08 UTC] No.43467119[source]▶
Sorry, I don't quite see what is wrong here.
replies(1): >>43467145 #
3. cplusplus6382 ◴[25 Mar 25 01:08 UTC] No.43467120[source]▶
Answer is correct no?
4. dangoodmanUT ◴[25 Mar 25 01:10 UTC] No.43467131[source]▶
9.9-9.11 =0.79

Might want to check your math? Seems right to me

5. kwakubiney ◴[25 Mar 25 01:10 UTC] No.43467132[source]▶
But the answer is correct? 9.9 is larger than 9.11
6. ◴[25 Mar 25 01:10 UTC] No.43467138[source]▶
7. manaskarekar ◴[25 Mar 25 01:11 UTC] No.43467145[source]▶
Parent is thinking Semantic Versioning.
replies(2): >>43467542 #>>43467670 #
8. AuryGlenz ◴[25 Mar 25 01:12 UTC] No.43467147[source]▶
I suggest we’ve already now passed what shall be dubbed the jschoe test ;)
replies(2): >>43467458 #>>43468092 #
9. ◴[25 Mar 25 01:12 UTC] No.43467148[source]▶
10. gaoryrt ◴[25 Mar 25 01:52 UTC] No.43467381[source]▶
This makes my day.
11. keyle ◴[25 Mar 25 01:58 UTC] No.43467419[source]▶
9.9 is larger than 9.11. This right here is the perfect example of the dunning-kruger effect.

Maybe try rephrase your question to "which version came later, 9.9 or 9.11".

12. bongodongobob ◴[25 Mar 25 02:03 UTC] No.43467451[source]▶
Lol, well I guess we've a achieved the functional equivalent of AGI, at least for you. Please don't delete your comment.
13. manaskarekar ◴[25 Mar 25 02:04 UTC] No.43467458[source]▶
jschoe's post is actually a Turing test for us. :)

(just kidding jschoe)

replies(1): >>43467639 #
14. aurareturn ◴[25 Mar 25 02:04 UTC] No.43467459[source]▶
+1 to Deepseek

-1 to humanity

replies(1): >>43473925 #
15. erichocean ◴[25 Mar 25 02:09 UTC] No.43467485[source]▶
This is hilarious, especially if it's unintentional.
replies(1): >>43467674 #
16. oefrha ◴[25 Mar 25 02:12 UTC] No.43467500[source]▶
I’ve legit seen a heated online debate with hundreds of comments about this question (maybe not the exact numbers), and I don’t think most participants were memeing. People are that bad at math. It’s depressing.
17. vbezhenar ◴[25 Mar 25 02:17 UTC] No.43467530[source]▶
But that’s correct. 9.9 = 9.90 > 9.11. Seems that it answered the question absolutely correctly.
replies(1): >>43467889 #
18. vbezhenar ◴[25 Mar 25 02:19 UTC] No.43467542{3}[source]▶
Semantic version contains 3 numbers.
19. declan_roberts ◴[25 Mar 25 02:37 UTC] No.43467639{3}[source]▶
He's Poe's law testing us.
20. declan_roberts ◴[25 Mar 25 02:46 UTC] No.43467670{3}[source]▶
One of many pet peeves with semver
21. declan_roberts ◴[25 Mar 25 02:47 UTC] No.43467674[source]▶
Poe's law in effect.
22. owebmaster ◴[25 Mar 25 03:00 UTC] No.43467732[source]▶
> I don't think anything else needs to be said here.

Will this humbling moment change your opinion?

23. sejje ◴[25 Mar 25 03:11 UTC] No.43467781[source]▶
What do you think the answer is?
replies(1): >>43468331 #
24. javchz ◴[25 Mar 25 03:35 UTC] No.43467889[source]▶
He's using Semantic versioning/s
25. sebastiennight ◴[25 Mar 25 04:19 UTC] No.43468092[source]▶
I will now refer to this as the jschoe test in my writing and publications as well!

It's interesting to think that maybe one of the most realistic consequences of reaching artificial superintelligence will be when its answers start wildly diverging from human expectations and we think it's being "increasingly wrong".

26. 7734128 ◴[25 Mar 25 05:23 UTC] No.43468331[source]▶
16 is obviously larger than both 9.9 and 9.11. AI will never be capable of thinking outside the box like that and find the correct answer.
27. WithinReason ◴[25 Mar 25 07:59 UTC] No.43468916[source]▶
You just failed the Turing test, now we know you're an LLM.
28. yencabulator ◴[25 Mar 25 17:43 UTC] No.43473925[source]▶
Based on the presented reasoning, that means humanity wins! Yay!