Qwen3.8 Max

Alibaba's flagship multimodal model is designed for coding, professional document analysis and extended tasks using application-supplied tools. Its 2.4-trillion-parameter mixture-of-experts architecture combines language processing with native image and video understanding.

CONTEXT256K

MAX OUTPUT—

LAST SCAN10 Oct 2026

🏆First observed free at GMI Cloud on 09 Oct 2026.

Text

Free aggregator routes

These are the aggregator routes recorded in the current or last published catalog. Each route keeps its exact model ID and thinking information.

Limits above: GMI Cloud. Available gateway routes are listed below.

GMI's official console lists Qwen3.8 Max separately from Qwen3.8 Max 0902. The supplied Qwen/Qwen3.8-Max route corresponds to Alibaba's advertised model documented at https://www.alibabacloud.com/help/en/model-studio/qwen3-8-max; no dated serving revision is established. It is not merged with Flash, 27B, the 0902 snapshot or the published base checkpoint.

GMI Cloud

Info

Thinking: Not documented

Model ID · GMI Cloud APIQwen/Qwen3.8-Max

Model sources

Qwen creator card distinguishing published weights from hosted Max

Use it through LiteLLM

Import the current LiteLLM-compatible YAML and use the exact model ID shown above. The catalog is generated from the same reviewed scan data as the download.

Download LiteLLM YAML