2026-02-17 CometAPI Adds Qwen3.5 Series!

[tensornull]

Qwen3.5 Series

qwen3.5-397b-a17b: The Qwen3.5 series 397B-A17B native vision-language model is based on a hybrid architecture design, fusing linear attention mechanisms with sparse Mixture-of-Experts (MoE) models to achieve higher inference efficiency. In various tasks such as language understanding, logical reasoning, code generation, agent tasks, image understanding, video understanding, and graphical user interfaces (GUI), it demonstrates excellent performance comparable to current top-tier frontier models. It possesses powerful code generation and agent capabilities, with good generalization for various agent scenarios.

qwen3.5-plus, qwen3.5-plus-2026-02-15, qwen3.5-plus-thinking: The Qwen3.5 native vision-language series Plus model is based on a hybrid architecture design, fusing linear attention mechanisms with sparse Mixture-of-Experts (MoE) models to achieve higher inference efficiency. In multiple task evaluations, the 3.5 series demonstrates excellent performance comparable to current top-tier frontier models, achieving a leap in performance in both pure text and multimodal aspects compared to the 3 series. This version is a snapshot from February 15, 2026.

Developer Documentation

[Quick Integration Guide (OpenAI Format)](https://apidoc.cometapi.com/chat)