Cyberax AI Playbook
cyberax.com
Model · Alibaba · Text

Qwen 3.5 122B

Alibaba's flagship open-weight model from the Qwen 3.5 generation. MoE architecture with 122B total / 10B active parameters; native multimodal (text + vision).

Modality
Text
License
Qwen License (Open weights)
Parameter size
122B (10B active)
Released
April 24, 2026
Last verified
May 10, 2026
Runs locally
Yes
Also handles
vision

Strengths

  • Strong multilingual and Chinese-language performance
  • MoE architecture — fast inference for the parameter count
  • Native vision capability
  • Strong coding and math performance

Weaknesses

  • License restricts commercial use at >100M MAU
  • Needs serious GPU footprint for self-hosting

Try it

WhereTypeNotes
Hugging Face weights Qwen License (FP8 variant linked; other quantizations available)
Ollama local Quantized variants available

Change log

  • — Initial entry. Sighted on Hugging Face (lastModified 2026-04-24).