3 ms·

Wasn't this release in Nov 2024 as a "preview" with similarly impressive performance? https://qwenlm.github.io/blog/qwq-32b-preview/ https://qwenlm.github.io/bl

Wasn't this release in Nov 2024 as a "preview" with similarly impressive performance? https://qwenlm.github.io/blog/qwq-32b-preview/ https://qwenlm.github.io/blog/qwq-32b-preview/

yorwba 2y ago
The benchmark scores in the new announcement are significantly higher than for the preview model.
- samus 2y ago
  That's good news, I was highly impressed already by what that model could do, even under heavy quantization.