Alibaba’s Qwen Deep Research Creates Live Webpages, Podcasts in Seconds

Qwen, the devoted AI analysis group throughout the Chinese language tech large Alibaba, launched a big improve to its AI chatbot final week, enabling customers to generate complete analysis paperwork on any subject.

You may then simply convert these paperwork into clear webpages or multi-speaker podcasts with only a few clicks.

Qwen Chat is much like ChatGPT, DeepSeek, or Claude by way of UI and is obtainable worldwide at no cost.

The brand new performance runs on three open-source fashions working in live performance: Qwen3-Coder handles net construction, Qwen-Picture generates inline graphics, and Qwen3-TTS powers dynamic audio narration.

Regardless of counting on open-source fashions, the end-to-end expertise—together with analysis execution, net deployment, and audio technology—is hosted and operated by Qwen as a managed service.

The workflow begins inside Qwen Chat, the place customers pose analysis questions. The AI conducts net searches after some clarifications, analyzes knowledge from public sources, and generates a complete report with citations.

From there, two new choices seem: “Net Dev” produces a reside, professional-grade webpage mechanically deployed and hosted by Qwen, full with inline graphics.

“Podcast,” in the meantime, presents an audio dialogue that includes dynamic multi-speaker narration, with 17 host voices and 7 co-host choices.

Testing the fashions

To evaluate how Qwen stacked up as a analysis instrument, we ran the identical advanced analysis question throughout it, Gemini, ChatGPT, and Grok. The duty, which will be reviewed on our GitHub repo, was to research philosophical and scientific arguments for and towards God’s existence. Every mannequin generated a full analysis report. The analysis concerned 5 standards: accuracy of claims and citations, data offered, readability of rationalization, mental richness, and general high quality.

TL;DR: Qwen Deep Analysis wins for analytical depth, quotation, and its distinctive auto-generated webpages, making it supreme for lecturers and creators. It is also the most effective all-in-one free various for researchers. However Gemini nonetheless leads in audio and video high quality, whereas ChatGPT and Grok stay effective for informal use however lack Qwen’s attain and Google’s polish.

This is a extra in-depth evaluation:

Accuracy: Had been philosophical positions and scientific claims represented accurately, with correct supply attribution?

Qwen nailed the main points. When discussing the cosmological argument, it correctly cited tutorial sources like Bertrand Russell’s “Why I’m not a Christian” and the controversy between William Lane Craig and Peter Atkins, with particular references. Not like different AI researchers like Perplexity’s or Grok, the vast majority of sources are respected and tutorial, typically even the Authentic Supply. It included hyperlinks from Stanford, Princeton, Oxford, Drew, however added pertinent evaluation from Quora and Fb when related.



Gemini matched this precision with 94 numbered citations, a few of which have been duplicated when referenced in numerous elements of the report.

It accurately distinguished between ideas. Each averted sloppy errors, equivalent to conflating biblical literalism with common theism.

ChatGPT relied closely on the Stanford Encyclopedia of Philosophy, however typically oversimplified. Grok gave correct summaries however with vaguer attribution—saying issues like “traced to Plato, Aristotle” with out particular works.

Consequence: Qwen and Gemini have been the most effective.

Data Offered: How thorough was the analysis?

Qwen was the one mannequin to incorporate a bit known as “Critiques of Atheism: The Burden of Proof and the Nature of Proof.” This part examined a kind of debate not one of the others touched. It distinguished between “weak atheism” (skepticism towards God claims) and “gnostic atheism” (constructive assertion God would not exist), and cited particular atheist thinkers like Gary Whittenberger’s “past an inexpensive doubt” customary.

This is an instance passage from Qwen: “Probably the most contentious points is the burden of proof. Bertrand Russell famously illustrated this along with his teapot analogy: simply as he couldn’t show {that a} tiny teapot doesn’t orbit the solar between Earth and Mars, he argued that theists couldn’t show that God does exist.”

No different mannequin went this deep into burden-of-proof debates as a result of it in all probability was not central to the subject. Gemini got here shut with robust protection of consciousness arguments and the “God-of-the-gaps” critique. ChatGPT included pragmatic arguments like Pascal’s Wager and explored real-world implications for ethics and coverage. Grok saved it concise—about one-third the size of Qwen’s report—however added a useful abstract desk.

Consequence: Qwen was probably the most exhaustive.

Readability: How was the analysis expressed?

Grok used a clear desk to prepare arguments by sort (Philosophical vs. Scientific, For vs. In opposition to). Its part breaks have been express: “Philosophical Arguments,” “Scientific Arguments,” “Sudden Element.” Anybody might scan it rapidly.

ChatGPT used tons of parenthetical clarifications, making advanced concepts extra digestible. Instance: “if God’s existence is even doable (i.e., logically coherent), then God exists essentially.” The “(i.e., logically coherent)” helps readers who aren’t philosophy majors.

Qwen and Gemini, then again, have been extra tutorial of their type. Qwen organized the content material below formal headings like “Theistic Arguments for God’s Existence: Cosmological and Teleological Foundations,” which made the entire studying really feel very dense, regardless of its accuracy. Gemini used Roman numerals (I. Introduction, II. Philosophical Arguments), which appeared structured however required nearer studying.

Each Qwen and Gemini goal researchers doing severe work. ChatGPT and Grok goal broader audiences.

Consequence: ChatGPT offered data probably the most clearly, adopted by Grok.

Variety of sources: Does the analysis draw from diverse traditions, disciplines, and views?

Qwen built-in technical philosophy (kalām, PSR, modal S5 logic) with reside scientific debates (Massive Bang singularities, quantum fluctuations, DNA performance). It defined issues, ensuring to be particular and provides background examples on positions and arguments.

For example, when explaining theistic arguments for God’s existence, Qwen constructed a desk to make it simpler to grasp the premises, critiques, and proponents of probably the most related arguments.

Gemini matched this by protecting consciousness arguments that the majority fashions ignored. It additionally warned towards “God-of-the-gaps” reasoning extra explicitly than rivals.

ChatGPT introduced distinctive worth with its large “Implications” part, exploring how the controversy shapes science training coverage, bioethics legal guidelines, and private attitudes towards loss of life. This was much less tutorial and extra pragmatic, however nonetheless related to understand the character of the investigation.

Grok lined the main arguments however with much less element. It talked about fine-tuning and the anthropic precept, however did not cite particular values or focus on issues too deeply.

Consequence: Qwen and Gemini have been the most effective.

High quality: Taking all collectively—rigor, coherence, scholarly worth—which analysis would you need to cite?

Each Qwen and Gemini produced experiences you can undergo your professor. Qwen’s distinctive power was balancing depth on each theistic traces and atheistic critiques, together with that burden-of-proof part. Gemini’s power was integrating scientific frontiers (consciousness, evolution, cosmology) with philosophical arguments.

ChatGPT delivered substantial pedagogical worth—nice for educating or understanding implications. Grok labored as a dependable primer or fast reference.

In different phrases, ChatGPT and Grok are in all probability those you’d use if you happen to simply need to know one thing rapidly for a dialog, to impress your nerd date, or refresh your data earlier than a presentation on one thing you already know

Ultimate Scores:

Qwen: 9/10
Gemini: 9/10
ChatGPT: 8/10
Grok: 6/10

The podcast battle: Qwen vs Gemini

Qwen’s podcast characteristic places it head-to-head with Google’s NotebookLM and Gemini, which pioneered AI-generated Audio Overviews.

Not like Gemini, Qwen presents a big number of host voices to select from. The construction is stable: two AI hosts have an precise dialog about your analysis, not only a text-to-speech read-through.

That stated, the voice high quality is inconsistent. Some voices are pure, however most of them sound robotic with bizarre accents. Throughout testing, one of many male hosts saved saying “oh oh oh” repeatedly, as a result of he was impressed. My spouse handed by and requested if I used to be watching porn.

With some trial and error, you could find an honest voice that works easily, and the standard will increase significantly.

However Gemini and NotebookLM crush Qwen right here. Google’s Audio Overviews characteristic—launched in NotebookLM in September 2024, expanded to Gemini in March 2025—sounds remarkably human. The speech patterns are pure, with back-and-forth banter and even humor.

Gemini’s podcasts really feel human and extra participating.

Gemini additionally presents video technology, which is a big benefit for individuals who favor an audiovisual method to understanding a subject moderately than studying lengthy chunks of textual content.

Qwen can’t do that—in truth, no different mannequin can.

If you would like full multimedia, together with audio, video, and net, Gemini is probably the most full bundle.

The webpage benefit

Past analysis high quality, Qwen’s killer characteristic is the auto-generated webpage. No different mannequin does this.

After your analysis finishes, you possibly can flip it right into a reside, hosted web site. Not a PDF or a Google Doc—an actual webpage with headers, formatted tables, embedded citations as hyperlinks.

The UI seems to be like Kimi; it options clear typography, responsive design and is immediately shareable.

ChatGPT customers have to repeat and paste into web site builders.

Gemini retains every part in Docs. Grok spits out textual content. Solely Qwen mechanically generates web-ready output.

That workflow benefit is good to have.