Standard & Paid Price
| Meter | Standard Price |
|---|---|
| Input price | $2.0000 / 1M tokens |
| Completion price | $6.0000 / 1M tokens |
| Cache read price | $0.2500 / 1M tokens |
| Cache creation price | $2.5000 / 1M tokens |
Description
Alibaba Qwen3.8-Max-0902 — the 2026-09-02 snapshot of Qwen3.8-Max (vendor alias qwen3.8-max-2026-09-02), the 2.4-trillion-parameter MoE flagship, frozen so integrations can pin this exact build; the vendor cites deeper coding for complex engineering projects and long-horizon autonomous development. 1,000,000-token context window, up to 131,072 output tokens; accepts text, image, and video input and returns text; supports thinking mode, tool calling, and structured output. The dynamic qwen3.8-max name moves to newer builds as the vendor releases them — pin this name to stay on this one. Priced identically to qwen3.8-max.
Capabilities
- Input modalities
- text, image, video
- Output modalities
- text
- Tasks
- chat, reasoning
Decision evidence
Qwen3.8-Max-0902 is the upgraded 2026-09-02 snapshot of Alibaba's Qwen3.8-Max API model, with the same 1,000,000-token context window, 131,072 output tokens and text, image and video input, and thinking on by default. On AI Coding Daily's coding leaderboard it scored 48.70, above the original qwen3.8-max, and kzhu's test gave it 46–47/60 with weak SVG animations.
2026-09-25 · Markdown evidence · JSON data
API endpoints
POST /v1/chat/completions(openai)POST /v1/messages(anthropic)