Qwen on Mac

The best way to use
Qwen on your Mac

Fello AI is a native Mac app giving you Qwen 3.8 Max through secure US-based infrastructure, plus every other frontier model in one place. No browser tabs, no data concerns.

Download on the Mac App Store
Thinking
Take more time to reason through complex, multi-step problems.
Featured
GPT-6
Power Speed | OpenAI
Claude 5.5
Power Speed | Anthropic
Grok 4.7
Power Speed | xAI
Gemini 4 Argon
Power Speed | Google
Muse Spark 1.3
Power Speed | Meta
Perplexity
Power Speed | Perplexity
Other
GLM 5.3
Power Speed | Zhipu AI
Kimi K3
Power Speed | Moonshot AI
DeepSeek V4.1
Power Speed | DeepSeek
Qwen 3.8 Max
Power Speed | Alibaba
Mistral
Power Speed | Mistral
Featured
GPT Image 2.5
Power Speed | OpenAI
Nano Banana 2
Power Speed | Google
Other
Grok Imagine 2
Power Speed | xAI
Seedream 5
Power Speed | ByteDance
FLUX.2
Power Speed | Black Forest Labs
Muse Image
Power Speed | Meta
What is Qwen

What Qwen actually is

Qwen is the family of AI models built by Alibaba, one of China's largest technology companies, through its Alibaba Cloud division. In Chinese the brand is known as Tongyi Qianwen (通义千问). Over the past few years Qwen has grown from a research project into one of the most influential model families in the world, and Alibaba has kept releasing new versions at a rapid pace, with Qwen 3.8 Max (released August 2026) as the current flagship.

Qwen comes in two flavors. The hosted "Max" models, like Qwen 3.8 Max, are Alibaba's most capable and are served through Qwen Cloud, Alibaba's API (this is the Qwen you get inside Fello AI). Alongside them, Alibaba publishes open weights: smaller Qwen models such as Qwen3.8-27B under the permissive Apache 2.0 license, and, for the first time with Qwen3.8, a Max-class model, the 2.4-trillion-parameter Qwen3.8-2.4T-A95B, under its own Qwen3.8-Max License. Those open releases, published on Hugging Face, ModelScope, and Kaggle, have made Qwen the world's most-downloaded and most-derived open model family, with hundreds of thousands of fine-tuned models built on top of it.

For everyday work, Qwen is a strong all-rounder. Alibaba has listed support for 201 languages and dialects since Qwen3.5, the foundation Qwen3.8 is built on, which makes it one of the most capable multilingual models available, and dedicated Qwen-Coder variants make it a genuinely good coding assistant. A hybrid "thinking / non-thinking" mode lets it toggle step-by-step reasoning on or off depending on the task.

Qwen vs Qwen 3.8 Max: Qwen is the model family (Tongyi Qianwen) from Alibaba. Qwen 3.8 Max is the current flagship, the hosted version you use through Alibaba's Qwen Cloud API and Qwen Studio. It is built on Qwen3.8-2.4T-A95B, whose weights Alibaba released in August 2026 under its own Qwen3.8-Max License; the hosted Max adds vision input, a 1M-token default context, and built-in tools. Smaller open Qwen models such as Qwen3.8-27B ship under the Apache 2.0 license, so anyone can download, modify, and self-host those. In Fello AI, Qwen queries are routed through secure US-based servers, not through Alibaba's own infrastructure.

201 languages
Listed by Alibaba since Qwen3.5
Apache 2.0
Open Qwen models are free to self-host
#1 open family
Most-downloaded and most-derived
Laptop to data center
A model for every size
Getting started

3 ways to use Qwen on Mac

01
Fello AI Recommended

A native Mac app (30-50MB, Apple silicon) that gives you Qwen 3.8 Max alongside Claude 5.5, GPT-6, Gemini 4 Argon, Kimi K3, Perplexity, and more. One subscription at $9.99/month. Works on Mac, iPhone, and iPad. Free tier available. Your queries are routed through a secure US-based pipeline, not Alibaba's own servers.

Download Fello AI
02
Qwen Studio and the Qwen apps

Use Qwen Studio at chat.qwen.ai in any browser, or Alibaba's official Qwen apps for iPhone, Android, macOS, and Windows. Free with usage limits and easy to try with no setup. Alibaba also makes QwenWork, a separate desktop agent for documents, slides, and web pages that needs macOS 14 or later. Both give you Qwen models only. Important caveat: your conversations are sent to and stored on Alibaba's servers in China, subject to Chinese data laws. Fine for casual queries, worth considering for sensitive work.

03
API and open weights

Developers can call Qwen through Qwen Cloud, Alibaba's official API service, where Qwen 3.8 Max costs $2 per million input tokens and $6 per million output tokens. Separately, you can download open Qwen weights from Hugging Face, ModelScope, or Kaggle and self-host them entirely on your own hardware: the smaller models under Apache 2.0, and the 2.4-trillion-parameter Max-class model under the Qwen3.8-Max License. Smaller sizes run comfortably on a modern Mac; the largest Mixture-of-Experts models need serious infrastructure.

Why Fello AI

Qwen, done right

Qwen is a strong model for math and coding, and Alibaba's own apps give you Qwen and nothing else. Fello AI gives you Qwen in a simple native Mac app, next to Claude, ChatGPT, and Gemini. You can also generate images with several AI image models and turn any answer into a real PowerPoint, Excel, or Word file.

Your data stays out of China
Fello AI routes Qwen queries through a secure US-based pipeline. Your conversations are not sent to or stored on Alibaba's own servers in China, making it the privacy-conscious way to access Qwen's capabilities.
One light app, every model
Alibaba's Qwen desktop app and its QwenWork agent only run Qwen models, and QwenWork alone takes about 700MB. Fello AI is a native app weighing 30-50MB that launches instantly, uses negligible RAM, and runs Qwen alongside every other model at a better price on any Mac running macOS 12 or later.
Qwen plus the frontier stack
Qwen 3.8 Max is a superb multilingual all-rounder with strong coding. For polished writing Claude 5.5 is stronger, and GPT-6 handles computer-use tasks best. Fello AI gives you all of them in one app with instant switching.
Side by side

Fello AI vs Qwen Studio

Feature Fello AI Qwen (chat.qwen.ai)
Price Free tier + $9.99/mo or $79.99/yr Free chat + paid usage-based API
Access to Qwen 3.8 Max ✓ Yes ✓ Yes (limited free tier)
Native Mac app ✓ macOS 12+, Intel + Apple silicon ✓ Official Qwen app (Qwen models only)
Data privacy ✓ US-based infrastructure ✗ Servers in China
Models available ✓ Qwen 3.8 Max, Claude 5.5, GPT-6, Gemini 4 Argon, Grok 4.7 and more ✗ Qwen only
Image generation ✓ GPT Image 2.5, Nano Banana 2, Seedream 5, FLUX.2 ✓ Qwen's own image models
PDF support ✓ Up to 16 PDFs at once Limited
iPhone and iPad ✓ Yes, same app Official Qwen app
In depth

A closer look at Qwen 3.8 Max

The world's most-derived open model family

Qwen is the most-downloaded and most-derived open model family in the world. Hundreds of thousands of fine-tuned and derivative models have been built on top of it, making it the default open-weight base for the global open-source community and a foundation that countless products quietly run on.

Multilingual across 201 languages

Alibaba lists support for 201 languages and dialects since Qwen3.5, the base Qwen3.8 is built on, which makes Qwen one of the strongest multilingual models available. Whether you're working in English, Chinese, Spanish, Arabic, or a less common language, Qwen tends to hold quality where many Western-centric models fall off.

Open weights, now including a Max-class model

Qwen3.8 is the first generation in which Alibaba released a Max-class model as open weights: Qwen3.8-2.4T-A95B, with 2.4 trillion parameters and 95 billion active, under a Qwen3.8-Max License that allows commercial use and asks only the largest AI service companies for a separate license. The hosted Qwen 3.8 Max adds vision input, a 1M-token default context, and built-in tools on top. Smaller models such as Qwen3.8-27B ship under Apache 2.0. Unlike GPT-6 and Claude, which are only available through company APIs, open Qwen models can be downloaded, inspected, modified, and self-hosted.

Your data stays private in Fello AI

Using Qwen through its own web app or Alibaba's API sends your conversations to servers in China. In Fello AI, your Qwen queries go through US-based infrastructure instead, which means your data never touches Alibaba's own servers.

Model comparison

How does Qwen compare to other AIs?

The AI landscape in 2026 is competitive at the frontier level. ChatGPT leads in computer-use automation, Claude 5 excels at writing and complex analysis, and Gemini pairs speed with a 1M-token window. Where does Qwen 3.8 Max actually stand?

Qwen 3.8 Max is Alibaba's latest flagship, an exceptional multilingual all-rounder with strong coding. Alibaba also publishes open weights, from Apache 2.0 models such as Qwen3.8-27B up to the 2.4-trillion-parameter Max-class model, which has made Qwen the most widely-adopted open model family in the world. For teams that value openness, self-hosting, or broad language coverage, Qwen is a genuinely compelling alternative to closed models.

Qwen VS ChatGPT

Qwen vs ChatGPT

ChatGPT is the established standard. Qwen is the open alternative.

GPT-6, which OpenAI runs in ChatGPT Work and Codex, is the strongest model for computer-use tasks. It can operate software, navigate interfaces, and complete multi-step desktop workflows autonomously. It is also backed by OpenAI's fully managed infrastructure, extensive ecosystem integrations, and consistent production reliability.

Qwen 3.8 Max is unusually strong at multilingual work, with reasoning and coding that are competitive at the task level. And for teams that want to inspect, audit, or self-host, Alibaba publishes open Qwen weights, since Qwen3.8 including a Max-class model, making Qwen one of the few frontier-level families that allows it.

Use ChatGPT for computer-use tasks, ecosystem integrations, and production reliability. Use Qwen when openness, multilingual reach, or self-hosting matters.

Qwen VS Claude

Qwen vs Claude

Claude is the quality benchmark. Qwen is the open one.

Claude 5 is the best model available for writing quality and following complex instructions. It produces the most natural, well-structured output of any major model, maintains consistency across very long documents, and is significantly more reliable at detailed multi-part tasks without drift.

Qwen 3.8 Max is competitive on many reasoning and coding tasks and wins decisively on language breadth, while Alibaba's open Qwen models come in sizes you can actually run yourself. Dedicated Qwen-Coder variants make it a strong coding assistant. What it does not match is Claude's polish on nuanced writing and instruction-following.

Use Claude 5 for writing quality, structured analysis, and complex instruction-following. Use Qwen for open-weight, multilingual, and self-hosted technical work.

Qwen VS Gemini

Qwen vs Gemini

Gemini scales wider. Qwen runs anywhere.

Gemini 3.8 Flash matches Qwen 3.8 Max's 1-million-token window and adds deep Google ecosystem integration. It processes enormous documents in a single session and pairs that with native Google Search grounding for real-time factual access. It also understands images, audio, and video natively.

Qwen 3.8 Max is exceptionally multilingual, and Alibaba publishes open Qwen weights you can self-host, in sizes from 27 billion to 2.4 trillion parameters. For organizations that cannot send data to Google's servers, or that want full control over the model, Qwen is a viable large-scale alternative.

Use Gemini for massive documents, Google integrations, and multimodal inputs. Use Qwen for open-weight, multilingual work where self-hosting matters.

Qwen VS Grok

Qwen vs Grok

Grok is real-time. Qwen is open.

Grok 4.7 is built around real-time information. It indexes X and the broader web live, making it the best model for breaking news, current sentiment, and anything time-sensitive. It is also more openly opinionated and willing to engage with edgy or controversial topics.

Qwen 3.8 Max is deeply multilingual, and Alibaba's open Qwen models are available across a huge range of sizes. For serious technical, coding, or multilingual work without a real-time data requirement, Qwen is the more flexible and self-hostable choice, and the more widely-adopted base to build on.

Use Grok for real-time information, live X data, and brainstorming. Use Qwen for open-weight, multilingual technical work where control matters.

Qwen VS Perplexity

Qwen vs Perplexity

Perplexity retrieves. Qwen builds.

Perplexity is a search-first tool. It retrieves and summarizes information from the web with citations, making it the best option when you want sourced answers quickly without doing manual research.

Qwen 3.8 Max is a general-purpose, multilingual reasoning model. Qwen Studio can search the web and run Deep Research, but search is one feature there, not the whole product, and Qwen is significantly better at writing, coding, complex reasoning, and turning information into something useful across many languages. These are complementary tools, not competing ones.

Use Perplexity to find and verify facts from the web. Use Qwen to reason, code, and build with that information.

Qwen VS DeepSeek GLM Kimi

Qwen vs other models

No single model wins every task.

Fello AI also gives you DeepSeek as a powerful, low-cost model, GLM as a strong open model, Kimi as a top open-weight model, and Muse Spark as Meta's cheap model for heavy coding. Each is a click away, right beside Qwen.

Use all of them in one app

No single model is best at everything. Qwen leads on openness and multilingual breadth, but Claude writes better, ChatGPT handles computer use, and Grok has live data. The most capable AI setups use multiple models routed to the right task. And through Fello AI, your Qwen queries run through US-based infrastructure, so your data never touches Alibaba's own servers. Fello AI puts Qwen, ChatGPT, Claude, Gemini, Grok, and Perplexity in one native app.

Metric Qwen 3.8 Max GPT-6 Claude 5 Gemini 3.8 Flash
Reasoning Excellent Excellent Excellent Very good
Coding Top tier Top tier Top tier Very good
Multilingual Best in class Very good Very good Good
Context window Up to 1M tokens 1M tokens 1M tokens 1M tokens
Open models available ✓ Yes, from Apache 2.0 models up to the 2.4T Max-class model ✗ No ✗ No ✗ No
Best for Open-weight, multilingual, self-hosting All-round work, operators Long-form writing, docs Speed, large docs, research
What Fello AI offers

What you can actually do in Fello AI

Switching between AI models in Fello AI
Use the right model for the task

Different models are better at different kinds of work. In Fello AI, you switch between ChatGPT, Claude, Gemini, Grok, and more without leaving the app or rebuilding your workflow. Compare outputs, move faster, and use the model that fits instead of forcing everything through one tool.

Fello AI supports PDF, Word, Excel, PowerPoint, images, and many more file formats
Chat with PDFs, images, and Office files

Upload PDFs, images, Excel sheets, slide decks, and documents, then ask for summaries, explanations, rewrites, or extracted insights. This makes Fello AI much more useful for study, research, reporting, and day-to-day professional tasks than a plain text chatbot.

Fello AI generating a downloadable PDF report from a spreadsheet
Create real documents you can download

Generate PowerPoint presentations, Excel spreadsheets with formulas, Word documents, and PDFs, then download and use them immediately. That turns AI from a brainstorming tool into something much closer to a real productivity workspace.

Fello AI web search results with cited sources
Search the web and keep working in one place

When you need current information, Fello AI searches the web with cited sources instead of forcing you to leave the app and manually piece things together. Especially useful when researching a topic, checking facts, or turning fresh information into a finished document.

Fello AI on Mac, iPhone, and iPad
One workflow across Mac, iPhone, and iPad

Fello AI is native on all Apple devices, so you keep the same app, the same models, and the same workflow whether you're on your Mac at a desk or on your phone on the go. No separate setup, no switching tools, no restarting the context.

For professionals

Fello AI gives professionals a practical way to use ChatGPT, Claude, Gemini, Grok, and other top models across Mac, iPhone, and iPad as part of one consistent workflow. Analyze PDFs, Word files, Excel sheets, presentations, and images, then turn the result into real downloadable documents for client work, internal workflows, and day-to-day execution.

For students

Fello AI helps students handle everyday academic work more efficiently. Summarize readings, explain difficult topics, compare answers across models, and work with PDFs, slides, notes, and study materials in one place. Instead of relying on a single AI answer, use the model that fits the task and move from quick explanations to deeper research without jumping between apps.

Common questions

Frequently asked questions

Yes. Alibaba lists official Qwen apps for macOS and Windows alongside iPhone and Android on qwen.ai, and it also makes QwenWork, a separate desktop agent for documents, slides, and web pages that needs macOS 14 or later. Both give you Qwen models only, and both run on Alibaba's own services. Several "Qwen" chat apps in the App Store come from third-party developers, not Alibaba. If you want Qwen next to Claude, GPT-6, Gemini, and Grok in one lightweight Mac app, with queries routed through US-based infrastructure, Fello AI is the simpler option.
The Qwen model itself is safe and capable. The data concern relates to Alibaba's hosted web app and API: your conversations are sent to servers in China and are subject to Chinese data laws, which require companies to cooperate with government data requests. For casual or non-sensitive queries this is unlikely to matter, but for confidential business, legal, or personal data it is worth considering. You can keep data fully local by self-hosting the open weights, or use Fello AI, which routes Qwen queries through US-based infrastructure.
Mostly. Alibaba open-sources many Qwen models under the permissive Apache 2.0 license, which allows free commercial use, modification, and self-hosting. Those open weights are published on Hugging Face, ModelScope, and Kaggle, and their wide range of sizes has made Qwen the most-downloaded and most-derived open model family in the world. With Qwen3.8, Alibaba also released a Max-class model's weights for the first time: Qwen3.8-2.4T-A95B, under its own Qwen3.8-Max License, which allows commercial use but asks the largest AI service companies for a separate license. The hosted Qwen 3.8 Max you reach through Qwen Cloud adds vision, a 1M-token default context, and built-in tools on top of those weights.
Very good. Alibaba ships dedicated Qwen-Coder models tuned specifically for programming, which handle code generation, debugging, and multi-file tasks strongly and rank among the better open-weight coding models. The hybrid "thinking" mode lets Qwen reason step by step through harder problems when needed. In Fello AI you can use Qwen for coding work and switch to Claude 5.5 or GPT-6 when you want a second opinion.
Alibaba lists 201 languages and dialects for Qwen, a figure it has given since Qwen3.5, the base Qwen3.8 is built on. That makes Qwen one of the most capable multilingual model families available. It performs well not just in English and Chinese but across many languages where Western-centric models often lose quality. If your work spans multiple languages, Qwen is one of the strongest options you can reach in Fello AI.
There are several paths. Qwen Studio at chat.qwen.ai and the official Qwen apps are free with usage limits. The open Qwen models are free to download and self-host, so your only cost is your own hardware. For hosted access, Qwen Cloud charges $2 per million input tokens and $6 per million output tokens for Qwen 3.8 Max. Through Fello AI's subscription at $9.99/month you get Qwen alongside Claude, GPT-6, Gemini, Grok, and more without managing any API billing directly.
Yes, in some areas. Like all China-based models, Qwen's hosted versions decline to discuss topics that are politically sensitive in China, including subjects such as Tiananmen Square, Taiwan independence, and similar issues. For technical, scientific, coding, multilingual, and most general-purpose tasks this is not a factor. For journalism, political research, or work involving Chinese geopolitics, it is a real limitation. Because the weights are open, self-hosted or fine-tuned versions can behave differently.
Yes, using the open Qwen models, which come in a wide range of sizes. Smaller open Qwen models run comfortably on a modern Apple Silicon Mac using tools like Ollama or LM Studio, with no data leaving your machine. Mid-sized models need more RAM, and the largest open Mixture-of-Experts models require serious infrastructure. The Max-class Qwen3.8-2.4T-A95B weights are public too, but at 2.4 trillion parameters they need data-center hardware, not a Mac. This flexibility is a big part of why Qwen is so widely adopted. For most users, the cloud-based options (Fello AI or Qwen Studio) are more convenient.
Yes. Fello AI works on any Mac running macOS 12 Monterey or later, including Intel-based Macs. Because Fello AI uses cloud-based AI (API calls to Qwen, Anthropic, OpenAI, etc.), your local hardware performance does not affect the quality or speed of the AI responses. This is different from running Qwen locally, which for larger models requires Apple Silicon and plenty of RAM.
Fello AI provides access to Qwen 3.8 Max, Alibaba's latest and most capable flagship, which became generally available in August 2026. Fello AI keeps model access updated as new versions are released, so you are not locked into an older version, and Qwen queries are routed through secure US-based infrastructure rather than Alibaba's own servers.

All the AI you need.
One beautiful app.

Download Fello AI for Mac, iPhone, and iPad. Free to start.

Fello AI running on Mac, iPad, and iPhone
Download on the App Store

4.7 rating·27,000+ reviews·Free to start