DataLearner logo
QW

Qwen3.6-35B-A3B

Reasoning modelQwen3.6
Release date: 2026-04-16Updated: 2026-07-17 21:57:31.0273,240
Live demoGitHubHugging FaceCompare
Parameters
35B
Context length
200K
Chinese support
Supported
Reasoning ability

Qwen3.6-35B-A3B is a reasoning model from Alibaba, released on 2026-04-16. It accepts text, image, and video input and returns text output. The cataloged parameter count is 35B, with 3B active parameters per inference. The recorded context window is 200K, and the recorded maximum output is 80K. Cataloged capabilities include Reasoning model and Multilingual. The checkpoint is listed under the Apache 2.0 license, with the license link included in the model references. Use the linked references to confirm current access, licensing, and provider-specific limits.

Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology

Qwen3.6-35B-A3B

Model basics

Reasoning traces
Supported
Thinking modes
Thinking modes not supported
Context length
200K tokens
Max output length
80K tokens
Model type
Reasoning model
Modality (in / out)
Text, Image, Video → Text
Release date
2026-04-16
Model file size
No data
MoE architecture
Yes
Total params / Active params
35B / 3B
Knowledge cutoff
No data
Qwen3.6-35B-A3B

Open source & experience

Code license
Apache 2.0
Weights license
Apache 2.0- Commercial use permitted
GitHub repo
https://github.com/QwenLM/Qwen3
Hugging Face
https://huggingface.co/Qwen/Qwen3.6-35B-A3B
Live demo
https://chat.qwen.ai
Qwen3.6-35B-A3B

Official resources

Paper
Paper
DataLearnerAI blog
No blog post yet
Qwen3.6-35B-A3B

API details

API speed
4/5
No public API pricing yet.
Qwen3.6-35B-A3B

Benchmark Results

Qwen3.6-35B-A3B currently shows benchmark results led by GPQA (1 / 15, score 86), MMLU Pro (24 / 132, score 85.20), LiveCodeBench (28 / 123, score 80.40). This page also consolidates core specs, context limits, and API pricing so you can evaluate the model from benchmark results and deployment constraints together.

Thinking

General Knowledge

4 evaluations
Benchmark / mode
Score
Rank/total
C-Eval
Thinking Mode
90
7 / 10
GPQA
Thinking Mode
86
1 / 15
MMLU Pro
Thinking Mode
85.20
24 / 132
HLE
Thinking Mode
21.40
112 / 170

Coding and Software Engineer

4 evaluations
Benchmark / mode
Score
Rank/total
LiveCodeBench
Thinking Mode
80.40
28 / 123
SWE-bench Verified
Thinking Mode
73.40
46 / 111
SWE-bench Multilingual
Thinking Mode
67.20
20 / 22
SWE-Bench Pro - Public
Thinking Mode
49.50
40 / 51

AI Agent - Tool Usage

2 evaluations
Benchmark / mode
Score
Rank/total
Terminal Bench 2.0
Thinking Mode
51.50
32 / 47
Tool Decathlon
Thinking Mode
26.90
9 / 9

Math and Reasoning

2 evaluations
Benchmark / mode
Score
Rank/total
AIME 2026
Thinking Mode
92.70
9 / 18
IMO-AnswerBench
Thinking Mode
78.90
19 / 21
View benchmark analysisCompare with other models

Compare with other models

No curated comparisons for this model yet.

Want a custom combination? Open the compare tool

Qwen3.6-35B-A3B

Publisher

Qwen3.6-35B-A3B

Model Overview

Qwen3.6-35B-A3B is a reasoning model from Alibaba, released on 2026-04-16.

It accepts text, image, and video input and produces text output. Its cataloged capabilities include Reasoning model and Multilingual. The cataloged parameter count is 35B, with 3B active parameters per inference. The recorded context window is 200K, and the recorded maximum output is 80K.

The checkpoint is listed under the Apache 2.0 license, with the license link included in the model references. The evaluation section contains 12 cataloged benchmark results with their recorded modes and scores. The page links 4 release, model-card, repository, or provider references for checking the underlying claims. Specifications, availability, and prices can change; undisclosed values are intentionally left unstated.

Qwen3.6-35B-A3B

FAQ

What is Qwen3.6-35B-A3B?

Qwen3.6-35B-A3B is a reasoning model from Alibaba, released on 2026-04-16. It accepts text, image, and video input and returns text output. The cataloged parameter count is 35B, with 3B active parameters per inference. The recorded context window is 200K, and the recorded maximum output is 80K. Cataloged capabilities include Reasoning model and Multilingual. The checkpoint is listed under the Apache 2.0 license, with the license link included in the model references. Use the linked references to confirm current access, licensing, and provider-specific limits.

What input and output modalities does Qwen3.6-35B-A3B support?

The current model record lists text, image, and video as input and text as output.

What are the main recorded specifications for Qwen3.6-35B-A3B?

The cataloged parameter count is 35B, with 3B active parameters per inference. The recorded context window is 200K, and the recorded maximum output is 80K. Fields without a source-backed value remain undisclosed.

Why is no API price shown for Qwen3.6-35B-A3B?

Open-weight checkpoint; no first-party hosted price published for this exact checkpoint.

Are benchmark results available for Qwen3.6-35B-A3B?

The evaluation section contains 12 cataloged benchmark results with their recorded modes and scores. Compare only results that use the same benchmark version and evaluation mode.

Is Qwen3.6-35B-A3B open source?

The checkpoint is listed under the Apache 2.0 license, with the license link included in the model references. Review the linked license text before commercial or derivative use.

DataLearner on WeChat

Follow DataLearner on WeChat for AI model updates and research notes.

DataLearner WeChat QR code