Qwen 3.6 35B-A3B

Alibaba

Qwen 3.6 35B-A3B open-weight checkpoint: 35B MoE, 3B active/token. 262K context, Apache-2.0.

Text Generation Local Latest Qwen Family v3.6

Documentation Back to Models

Parameters

35B

params

Context Window

262K

tokens

Max Output

tokens

Input Price

per 1M tokens

Output Price

per 1M tokens

Qwen Family 14 models

The full Qwen line by generation — pricing and capabilities vary across the family.

3.6

3.6 35B-A3B Current Latest

Local inference, agentic workflows, coding

262K

context

Apr 2026

3.5

3.5 35B-A3B

Efficient inference, agentic tasks, edge deployment

Reasoning, coding, analysis, multimodal

AI agents, multilingual tasks, reasoning, multimodal

General-purpose, coding, reasoning, multilingual

Flagship Qwen 3, agent tasks

Advanced reasoning, agents

Hybrid reasoning, MoE architecture

2.5

Flagship Qwen 2.5, beats DeepSeek V3

Alibaba

Deep reasoning, chain-of-thought

Multilingual, coding, math

Complex tasks, analysis

Professional applications

Code generation, debugging

128K

context

Sep 2024

Capabilities

👁️

Vision

⚡

Function Calling

📋

JSON Mode

🌊

Streaming

💬

System Prompt

🖥️

Code Execution

🔍

Web Search

🔌

MCP Support

Local Model Specs

Architecture

MoE (3B active)

Details

Release Date: April 16, 2026
Knowledge Cutoff: -
Source: Local
License: Apache 2.0
Model ID: qwen3.6-35b-a3b

Last updated: May 27, 2026