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

(qwenlm.github.io)

544 points tosh | 1 comments | 24 Mar 25 18:35 UTC | HN request time: 0.231s | source

1. naasking ◴[25 Mar 25 13:51 UTC] No.43471298[source]▶

I found Qwen never completed answering my standard coding task that I ask to check a model. Claude did great, DeepSeek R1 did well.