Agentic Coding
Stronger at repo-level coding, terminal work, and multi-step developer tasks than the Qwen 3.5 hosted line.
Qwen3.6-Plus — Toward Real-World Agents
Qwen3.6-Plus is the latest Qwen release with improved coding, tool use, multimodal reasoning, and a 1M token context window. Try it free.
Ready To Chat
Online
ThinkingQwen3.6-Plus is ready
Qwen3.6-Plus is the default model for this page. Hosted Qwen 3.6 release for agentic coding, stronger tool use, and sharper multimodal reasoning.
Pick a model, decide whether this needs web search or thinking, then start with a real prompt.
General
Analysis
Starter prompts
Free to try in the browser. This page gives you the key context, benchmark numbers, and usage notes without the launch-post framing.
Generation
Qwen 3.6
Status
Released
Context
1M default
Access
Hosted API
Jump to another Qwen model
Overview
What Is Qwen 3.6?
Qwen 3.6 is the next step after Qwen 3.5, and Qwen3.6-Plus is the current hosted release in that line. It is aimed at agentic coding, tool use, multimodal reasoning, and longer-running repo tasks.
1M Default Context
Qwen3.6-Plus ships with a 1M default context window for longer documents, repos, and sessions.
Agentic Coding Focus
Repository-level coding, terminal work, and tool use sit at the center of the Qwen3.6-Plus upgrade.
preserve_thinking
The API adds preserve_thinking for longer multi-step agent workflows where you want reasoning to persist across turns.
Qwen3.6-Plus Benchmark
How Qwen3.6-Plus compares to nearby models in the Qwen family.
Model
MMLU-Pro
GPQA / GPQA-family
LiveCodeBench v6
Qwen3.5-397B-A17B
Flagship open-weight Qwen3.5 model, also the base model behind Qwen3.5-Plus.
Updated 2026-04-02
MMLU-Pro
87.8
GPQA / GPQA-family
88.4
LiveCodeBench v6
83.6
Qwen3.5-Plus
HostedHosted version built on Qwen3.5-397B-A17B with additional tooling and a 1M context window.
Scores reference the Qwen3.5-397B-A17B base model.
Updated 2026-04-02
MMLU-Pro
87.8
GPQA / GPQA-family
88.4
LiveCodeBench v6
83.6
Qwen3.6-Plus
NewCurrent Qwen 3.6 hosted release with agentic coding, stronger tool use, and multimodal reasoning.
1M default context window with preserve_thinking support.
Updated 2026-04-02
MMLU-Pro
88.5
GPQA / GPQA-family
90.4
LiveCodeBench v6
87.1
Scores are from public model cards and the qwen.ai release page. Hosted models are labeled with their open-weight base.
Updated 2026-04-02Use Cases
Why Try Qwen3.6-Plus Now
Use Qwen3.6-Plus when you want the current hosted Qwen model for coding, tool use, multimodal work, and long sessions.
Tool Use
A better fit for code agents, general agents, and tool-heavy workflows that need structured calls.
Multimodal Reasoning
Handles documents, OCR-style inputs, interface screenshots, and other multimodal tasks more comfortably than text-only workflows.
Long Context Work
The 1M default context window is aimed at longer sessions and larger real-world tasks, not just short demos.
CLI and Agent Workflows
Fits repo-level coding, terminal tasks, and tool-heavy automation flows.
Long Sessions
The 1M context window makes it easier to keep large docs, codebases, and long chats in one run.
FAQ
Qwen3.6-Plus FAQ
Common questions about the current Qwen 3.6 hosted release.
Is Qwen 3.6 better than Qwen 3.5?
Qwen3.6-Plus is stronger on agentic coding, tool use, multimodal reasoning, and long-context work. Qwen 3.5 still matters if you want the broader open-weight model family.
Should I switch from Qwen 3.5 to 3.6?
Switch when you specifically want the current hosted Qwen 3.6 release for agentic coding, tool use, or the 1M default context window. Stay on Qwen 3.5 when you need the broader public open-weight family today.
Does Qwen3.6-Plus support tool calling?
Yes. Tool use is one of its main strengths, especially for multi-step coding and agent workflows.
Is Qwen3.6-Plus free to try here?
Yes. You can try Qwen3.6-Plus on this site through the hosted browser route.
What is Qwen3.6-Plus's context window?
1M tokens by default, same as Flash and Plus.
Can I use Qwen3.6-Plus for coding?
Yes. Qwen3.6-Plus is specifically improved for agentic coding — multi-step code generation, debugging, and refactoring across files.
Is Qwen3.6-Plus open source?
No. Qwen3.6-Plus is a hosted-only model. If you need open weights, use the Qwen 3.5 models which are available under Apache 2.0.
Does Qwen3.6-Plus support images?
Yes. Qwen3.6-Plus handles multimodal input — text, images, and documents in the same conversation.
Is Qwen3.6-Plus good for long documents and repos?
Yes. The 1M default context window makes it a strong fit for large docs, long chat histories, and repo-scale coding tasks.