LLM Reference

Qwen3.6 Models by Alibaba

AlibabaApache 2.0Open source
6 models2026Up to 1m ctxFrom $0.15/1M input

Details

ResearcherAlibaba
LicenseApache 2.0OSI-approved
Commercial useCommercial use: permitted
Models6
Released2026
Max context1m

Capabilities

VisionAll models
MultimodalAll models
Reasoning2 of 6 models
Function Calling4 of 6 models
Tool Use4 of 6 models
Structured Outputs2 of 6 models

Links

Website

About

Qwen3.6 is a family of 6 AI models by Alibaba, released in 2026.

Current Variants

Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.

6 in view
Qwen3.6-27BCurrent

Use when the workload needs code, 262k context, and 27B parameters.

2026-04code262k context27B parameters
Qwen3.6 Max PreviewCurrent

Use when the workload needs 256k context, reasoning, and tool use.

2026-04256k contextreasoningtool use
Qwen3.6-35B-A3BCurrent

Use when the workload needs code, 262k context, and 35B parameters.

2026-04code262k context35B parameters
Qwen3.6-FlashCurrent

Use when the workload needs 1m context and multimodal inputs.

2026-041m contextmultimodal inputs
Qwen3.6-MaxCurrent

Use when the workload needs 262k context and multimodal inputs.

2026-04262k contextmultimodal inputs
Qwen3.6-PlusCurrent

Use when the workload needs code, 1m context, and tool use.

2026-04code1m contexttool use

Release Timeline

1 release group
2026-04
6 current
Qwen3.6 Max Preview
256k contextreasoningtool use
Current
Qwen3.6-27B
code262k context27B parameters
Current
Qwen3.6-35B-A3B
code262k context35B parameters
Current
Qwen3.6-Flash
1m contextmultimodal inputs
Current
Qwen3.6-Max
262k contextmultimodal inputs
Current
Qwen3.6-Plus
code1m contexttool use
Current

Specifications(6 models)

Qwen3.6 model specifications comparison
ModelReleasedContextParametersVisionMultimodalReasoningFn CallingTool UseStructured Outputs
Qwen3.6-27B2026-04262k27BYesYesYesYesYesNo
Qwen3.6 Max Preview2026-04256k—YesYesYesYesYesYes
Qwen3.6-35B-A3B2026-04262k35BYesYesNoYesYesNo
Qwen3.6-Flash2026-041m—YesYesNoNoNoNo
Qwen3.6-Max2026-04262k—YesYesNoNoNoNo
Qwen3.6-Plus2026-041m—YesYesNoYesYesYes

Pricing

Qwen3.6 model pricing by provider
ModelProviderInput / 1MOutput / 1MType
Qwen3.6-35B-A3BOpenRouter$0.15$1Serverless
Qwen3.6-FlashOpenRouter$0.1875$1.125Serverless
Qwen3.6-35B-A3BNovita AI$0.248$1.485Serverless
Qwen3.6-FlashAlibaba Cloud PAI-EAS$0.25$1.5Serverless
Qwen3.6-27BOpenRouter$0.32$3.2Serverless
Qwen3.6-27BAlibaba Cloud PAI-EAS$0.32$3.2Serverless
Qwen3.6-PlusAlibaba Cloud PAI-EAS$0.325$1.95Serverless
Qwen3.6-PlusOpenRouter$0.33$1.95Serverless
Qwen3.6-PlusVercel AI Gateway$0.5$3Serverless
Qwen3.6-27BVercel AI Gateway$0.6$3.6Serverless
Qwen3.6-27BNovita AI$0.6$3.6Serverless
Qwen3.6 Max PreviewOpenRouter$1.04$6.24Serverless
Qwen3.6 Max PreviewAlibaba Cloud PAI-EAS$1.04$6.24Serverless
Qwen3.6 Max PreviewVercel AI Gateway$1.3$7.8Serverless

Popular comparisons in this family

Frequently Asked Questions

What is Qwen3.6 used for?
Qwen3.6 is used for code, vision and multimodal work, and reasoning. The family description and listed model capabilities point to those workloads as the best fit.
How does Qwen3.6 compare to Tongyi DeepResearch?
Qwen3.6 by Alibaba is strongest where you need code, while Tongyi DeepResearch by Alibaba is the closest related family to check for adjacent model selection. Qwen3.6 has 6 listed variants and reaches up to 1m context, while Tongyi DeepResearch reaches up to 131k context, so compare the specs and pricing tables before choosing a production model.
Which Qwen3.6 model should I use?
For the lowest listed input price, start with Qwen3.6-35B-A3B through OpenRouter at $0.15/1M input tokens. For the most capable/latest local choice, evaluate Qwen3.6 Max Preview with 256k context and reasoning, tool use, function calling, structured outputs, and multimodal inputs.