1.10.0 v3.1.0
QwenLM/Qwen3-ASRSep 4, 2026 → Sep 3, 20260 days between releases
Generating AI comparison
Analyzing what changed between 1.10.0 and v3.1.0. This page updates automatically — usually within a minute.
The Two Releases
1.10.0
v1.10.0: Faster Training & Inference
Sep 4, 2026· 2 months ago
A major performance-focused release that reduces training time by 44% and inference speed by up to 46%. It introduces a GPU batched linear-assignment solver and alters several default behaviors.
Full release notesv3.1.0
v3.1.0
Sep 3, 2026· 2 months ago
Focuses on training improvements, new language models (Hindi and Korean), and documentation updates to support CPU-only usage.
Full release notes