Alibaba has 10 models tracked on this site, from Qwen 3 235B in 2025 to Qwen3.8-Omni-Flash, released Sep 17, 2026. Its highest-ranked model, Qwen3.8 Max, is #8 of 39 on the Singularity Index with a score of 52.0.
Newest first. Bars show each model's Singularity Index score out of 100. Provisional scores have partial benchmark coverage and are not ranked yet.
Accepts text, images, audio, and video with text-only output. Supports tool calling and a one-million-token context. Model Studio lists availability from September 17; the launch article followed September 20. Independent benchmark coverage is pending.
NOT YET SCOREDQwen efficiency tier at AA Intelligence Index 55.8, within striking distance of the 58.1 flagship.
Small open-weight Qwen 3.8 tier, holding GPQA 90.5 at 27B parameters.
Very large Qwen MoE (2.4T total, 95B active) posting the family's highest GPQA at 93.5.
Qwen flagship successor to 3.7 Max. AA Intelligence Index 58.1 with a large agentic jump (58.4), at GPQA 92.7.
Frontier-tier Qwen flagship. AA Intelligence Index 56.6, GPQA 92.3, Terminal-Bench 50.8 — closes the gap to Western frontier at lower cost.
Top-tier Qwen variant above Plus, highest Arena Elo in the family at 1511
Agentic coding focus, 1M context, strong multimodal performance
Open-weight 397B MoE with visual agentic capabilities, 201 languages
Hybrid reasoning MoE trained on 36T tokens across 119 languages