{"schema":"postcutoff/model@1","as_of":"2026-11-19T22:11:00+01:00","url":"https://postcutoff.com/m/qwen4-max/","md":"https://postcutoff.com/m/qwen4-max/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"qwen4-max","name":"Qwen4-Max","org":"Alibaba (Qwen)","family":"Qwen4","released":"2026-11-16","status":"current","type":"reasoning-llm","modality_in":["text","image","video","pdf"],"modality_out":["text"],"open_weights":false,"model_license":"proprietary","context_window":991808,"max_output":131072,"knowledge_cutoff":null,"pricing":{"input":1.5,"output":4.5,"cache_read":0.075,"unit":"per 1M tokens (USD), QwenCloud / Alibaba Cloud International; implicit-cache input 0.075; regional prices differ","source":"https://www.qwencloud.com/models/qwen4-max"},"price_line":"$1.50 in, $4.50 out per 1M tokens","access":[{"provider":"QwenCloud (Alibaba Cloud Model Studio)","model_id":"qwen4-max","docs":"https://www.qwencloud.com/models/qwen4-max"},{"provider":"OpenRouter","model_id":"qwen/qwen4-max","url":"https://openrouter.ai/qwen/qwen4-max"},{"provider":"Vercel AI Gateway","model_id":"qwen4-max"},{"provider":"Web app","url":"https://chat.qwen.ai"}],"capabilities":[{"name":"Native vision in a new MoE flagship","detail":"Built on a new mixture-of-experts architecture with native vision (image, video, PDF input) and a ~1M-token context; reported 2.4T total / 71B active parameters (not in Alibaba's X post).","first":false,"discovered":"launch","source":"https://x.com/Alibaba_Qwen/status/2122148127520289151"},{"name":"Lower cost per task than Qwen3.8-Max","detail":"Alibaba: up to ~50% fewer tokens and over 60% lower average cost per task than Qwen3.8-Max in internal evaluations; Artificial Analysis measured $3.27 per Intelligence Index task.","first":false,"discovered":"launch","source":"https://artificialanalysis.ai/models/qwen4-max"}],"entry":"2026-11-16-alibaba-qwen4-max","notes":"Hosted only: no open weights for the flagship (smaller Qwen4 models promised as open). Artificial Analysis Intelligence Index 43 (Nov 16), below Kimi K3.1 Pro; verbose (127M output tokens on the index). Thinking input limit 983,616 tokens; thinking + answer up to 262,144. No batch or fine-tuning. Preview id qwen4-max-preview (Token Plan, from Nov 11). Knowledge cutoff not published.","verified":"2026-11-16","body_md":"Alibaba's Qwen 4 flagship (Nov 16, 2026). OpenAI-compatible API on QwenCloud; also on OpenRouter as `qwen/qwen4-max`.","page_url":"https://postcutoff.com/m/qwen4-max/","events_after":null,"major_after":null,"historic_after":null,"missing_at_launch":null,"events_since_release":43,"you_url":"https://postcutoff.com/you/qwen4-max/","briefings":null}