TPSTokens Per Second

Qwen3.5-Max-Preview

Alibaba / Qwen🇨🇳 China
active

Alibaba's biggest Qwen3.5 model with over 1 trillion parameters. Supports thinking mode with 82K chain-of-thought output. Part of the Qwen3.5 series featuring hybrid linear attention + MoE architecture. Optimized for reasoning, agentic tasks, RAG, and tool calling.

Compare with other models →
Context window262K tokens
Input / 1M tokens$1.2
Output / 1M tokens$6

Version History

qwen3.5-max-preview-2026-03-18major

Qwen3.5-Max-Preview launches as Alibaba's largest model (1T+ params) in the Qwen3.5 series. 262K context with thinking mode (82K CoT). Beats previous flagship on reasoning, multilingual, and agentic tasks. $1.20/$6.00 per 1M tokens.