Alibaba· QwenNew model

Qwen1.5-32B: Fitting the Capstone of the Qwen1.5 Language Model Series

Alibaba announces a new model with approximately 30 billion parameters, aiming to balance performance, efficiency, and memory footprint.

From the source

A growing consensus within the field now points to a model with approximately 30 billion parameters as the optimal “sweet spot”

qwenlm.github.io

Announced

02 Apr 2024 05:33 UTC

Detected

03 Aug 2026 21:12 UTC (+853d after)