Qwen4-Max

Released

Qwen4-Max is an AI model released by Qwen on Wednesday, Nov 11 2026, 70 days after Qwen3.8-Max-0902. It has a 1M token context windowHow much information the model can work with in one request, measured in tokens — pieces of words or other data. This includes your question, attachments, conversation and reply. Limits can vary by model variant and provider.. Benchmark results (shown below) cover APEX-Agents, DeepSWE 1.1, Terminal-Bench 4.0, JobBench, Humanity's Last Exam, BioMysteryBench, and 5 more.

Benchmarks

Coding

DeepSWE 1.1Agentic coding — Artificial Analysis' independent test of deep, agentic software-engineering work — the AI has to plan and carry out substantial coding tasks end to end. (Version 1.1 of the test.) Higher is better.
75.7%
#5 of 43

Terminal & CLI

Terminal-Bench 4.0Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? Version 4.0 recalibrated how much time, CPU and memory each task gets, removed eight tasks and fixed nineteen, so fewer runs fail for reasons that have nothing to do with the model. Scores are not comparable with earlier versions. Higher is better.
48.5%
#15 of 37

Agentic & tool use

APEX-AgentsExpert agentic work — expert-level agentic work tasks (AI Productivity Index)
72.1%
#1
JobBenchProfessional tool use — Tests the AI on professional workplace tasks that require using real work tools — the kind of multi-step jobs an office worker handles. Higher is better.
68.5%
#3 of 13
Agents’ Last Exam v1.1Agentic computer use — averaged rubric score on the 122-task October 17, 2026 v1.1 set, using ALE-Claw with a five-hour limit; separate from earlier ALE scores
46.7%
#2 of 2
AutomationBench (Artificial Analysis)Business workflows — The same business-workflow test, run by Artificial Analysis rather than by the lab, and graded strict pass@1: a workflow counts only if the whole of it is right on the first attempt. Read it as a separate board from the lab-run AutomationBench scores. Higher is better.
72.5%
#2 of 3

Reasoning & science

Humanity's Last ExamMultidisciplinary reasoning — Humanity's Last Exam — extremely hard expert questions across many subjects, written so you can't just look up the answer. “No tools” means the AI answers on its own. Higher is better.
51.7%no tools
#8 of 49
BioMysteryBenchBiology — Real biology puzzles that human experts have managed to crack — can the AI reach the same answers? Higher is better.
85%human solved
#8 of 11

Knowledge work

AA-Briefcase v1.1Knowledge work — Artificial Analysis agentic office-work eval (Elo, v1.1 rating fit)
1686
#4 of 14

Healthcare

HealthBench ProfessionalHealth — Realistic health conversations graded against detailed rubrics written by physicians — can the AI respond the way a careful medical professional would? Higher is better.
60%
#3 of 5

Multimodal

ChartographyChart tasks — The same chart-centred test with no tools: the AI has to read each chart unaided. Scores run far lower than the with-tools version, so read the two as separate tests. Higher is better.
54.2%no tools
#8 of 16

Compare Qwen4-Max with

Qwen4-Max

About

Qwen4-Max was announced in preview on November 11, 2026 as the flagship of the Qwen4 generation, available first through Alibaba Cloud's Token Plan subscription and billed in Credits. Model Studio listed the preview as qwen4-max-preview, accepting text, images and video with a one-million-token context, thinking mode, function calling and structured output. It followed the hosted Qwen3.8-Max tier and its September 2 revision.

The general release followed on November 16, 2026, with Qwen calling it a new AI teammate for work, creativity and discovery. Qwen described a new mixture-of-experts design with built-in vision and kept the one-million-token context. Its launch claims centred on efficiency, measured by Qwen itself: it beat Qwen3.8-Max while using up to about half the tokens, average cost per task on Work360 fell from $4.43 to $1.67, and token use on Terminal-Bench 4.0 was down about 48 to 49 percent. Qwen's launch chart put it at 75.7 on DeepSWE v1.1, narrowly ahead of every model it was charted against, and 72.5 on Artificial Analysis's AutomationBench, while Kimi K3.1 led the chart's Terminal-Bench 4.0 panel and Claude Opus 5.5 its ProgramBench panel. At launch it placed sixth in Code Arena's WebDev ranking, 74 points above Qwen3.8-Max, and was up 7.5 percent net on Agent Arena.

Frequently asked questions

Qwen4-Max was released by Qwen on Wednesday, Nov 11 2026.

All Qwen releases

25 tracked