qwen capabilities
14 mapped capabilities, each graded and dated. This is the diagnosis — the migration guide is the cure.
Capabilities
Chat history
provisionalverified 2 days agoQwen Chat saves conversation history for signed-in users, making previous sessions accessible across devices and sessions.
Deep Research
provisionalverified 2 days agoQwen Chat's Deep Research mode conducts multi-step web investigations on a topic and produces a structured, cited report; outputs can optionally be converted to a hosted webpage or multi-speaker podcast.
Developer API (Alibaba Cloud Model Studio / DashScope)
provisionalverified 2 days agoDevelopers can access all Qwen models programmatically via the Alibaba Cloud Model Studio API, which supports both a native DashScope SDK and an OpenAI-compatible endpoint.
Free access and plan
provisionalverified 2 days agoQwen Chat is free to use for general users on web and mobile, with no subscription required to access core features including image generation, web search, thinking mode, and multimodal input.
Image editing
provisionalverified 2 days agoQwen Chat lets users upload an image and modify it via text instructions, including style transfer, object addition or removal, background changes, and bilingual text edits.
Image generation
provisionalverified 2 days agoQwen Chat can generate images from text prompts using Qwen-Image, a 20B MMDiT image foundation model with strong text rendering in both English and Chinese.
Image upload and vision understanding
provisionalverified 2 days agoUsers can upload images to Qwen Chat for visual question answering, OCR, chart reading, object recognition, document parsing, and image editing via the Qwen VLo preview.
Mobile apps
provisionalverified 2 days agoQwen Chat is available as a free native app on iOS and Android, with feature parity to the web interface plus mobile-exclusive capabilities like a floating panel on Android.
Model picker
provisionalverified 2 days agoQwen Chat lets users choose from multiple Qwen models in the interface, including the latest Qwen3.5 and Qwen3 series, trading off capability and speed.
Thinking / reasoning mode
provisionalverified 2 days agoQwen Chat exposes a hybrid thinking mode powered by Qwen3 that lets users toggle between extended step-by-step reasoning and fast non-thinking responses within the same conversation.
Video generation
provisionalverified 2 days agoQwen Chat includes video generation powered by Alibaba's Wan model family, allowing users to generate short video clips from text prompts or uploaded images.
Voice and omni multimodal
provisionalverified 2 days agoQwen Chat supports real-time voice conversation and omni-modal input (text, image, audio, video) via the Qwen-Omni model family, with streaming speech output in multiple languages.
Web Dev (code artifacts)
provisionalverified 2 days agoQwen Chat's Web Dev mode generates fully functional front-end websites (HTML, CSS, JavaScript) from a single natural language prompt, rendered inline as a live preview.