Meet the team that turns your ideas into work.

Discover EightBots
EIGHTBOTS
Contact salesSign up
AlibabaQwen3.6

Qwen3.6-35B-A3B

A sparse vision-language release with preserved reasoning context.

Practical use

Practical guidance. Consider this model when your serving stack supports sparse experts. Measure both resident memory and processing speed, since the active parameter count describes computation rather than the full weight footprint.

Specifications & access

Qwen3.6-35B-A3B specifications
DetailValue and conditions
AccessDownloadable
Model identifiersQwen/Qwen3.6-35B-A3B
Architecture35 billion total and three billion active parameters; 40 layers.Qwen3.6-35B-A3B model card
Context262,144 native tokens; the card documents extension to about one million.Qwen3.6-35B-A3B model card
LicenceApache 2.0.Qwen3.6-35B-A3B model card

Related models

  • Qwen3.6-27B

    Alibaba ยท Qwen3.6

    A dense vision-language release with preserved thinking context.

    Launch date not confirmedDownloadable