←back to thread

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

(qwenlm.github.io)
544 points tosh | 3 comments | 24 Mar 25 18:35 UTC | HN request time: 0s | source
Show context
101008 ◴[24 Mar 25 20:15 UTC] No.43464981[source]▶
Silly question: how can OpenAI, Claude and all, have a valuation so large considering all the open source models? Not saying they will disappear or be tiny (closed models), but why so so so valuable?
replies(10): >>43465120 #>>43465472 #>>43465899 #>>43465934 #>>43466072 #>>43466079 #>>43467117 #>>43467174 #>>43467244 #>>43468107 #
elorant ◴[24 Mar 25 22:32 UTC] No.43466079[source]▶
The average user won't self-host a model.
replies(2): >>43466158 #>>43467849 #
epolanski ◴[24 Mar 25 22:42 UTC] No.43466158[source]▶
...yet
replies(2): >>43468526 #>>43475614 #
1. 8n4vidtmkvmk ◴[25 Mar 25 06:22 UTC] No.43468526{3}[source]▶
I'm not sure how it'll ever make sense unless you need a lot of customizations or care a lot about data leaks.

For small guys and everyone else.. it'll probably be cost neutral to keep paying OpenAi, Google etc directly rather than paying some cloud provider to host an at best on-par model at equivalent prices.

replies(2): >>43472086 #>>43474088 #
2. ◴[25 Mar 25 14:53 UTC] No.43472086[source]▶
3. mdp2021 ◴[25 Mar 25 18:03 UTC] No.43474088[source]▶
> unless you need a lot of customizations or care a lot about data leaks

And both those needs are very normal. "Customization" in this case can just be "specializing the LLM on local material for specialized responses".