qwen3.5:cloud

Upstream reference: Qwen3.5-397B-A17B

Publisher: Qwen Team, Alibaba Cloud

This vision-language model combines a hybrid attention layout with a sparse mixture-of-experts language component.

Publisher-described context, separate from Sigma5C observations and recorded catalog statistics.

Published upstream specifications

Language component total parameters

397 billion Source

Language component active parameters

17 billion Source

Native context

262,144 tokens Source

Upstream license label

Apache-2.0 Source

Publisher-described uses

• Text/visual generation and tool-oriented applications described by Qwen. Source

Reference limitations

• The card distinguishes native context from extended configuration and the separate hosted Qwen3.5-Plus product. A cloud alias does not establish either product’s exact serving settings. Source

A recorded tag and an upstream card association do not verify the deployed checkpoint, quantization, configured context or current hosted availability.

Publisher sources

• Qwen3.5-397B-A17B official documentation — Qwen Team, Alibaba Cloud. Checked 2026-11-02.

Links

Recorded references do not establish downloadable weights, licensing or current availability.

• Alibaba Cloud Qwen — official site
• Ollama
• Hugging Face

Where qwen3.5:cloud comes from

Named in 3 observation pages

• Academic observation — 32 cloud models × Cline and Pi · V1
• Aider — 169 runs across 83 models · V1
• Function-calling — 9 cloud models × 30 prompts · V1