Alibaba Cloud Model Studio offers Qwen and third-party models for text, image, audio, and video.
Text generation
 qwen3.8-max | qwen3.7-plus | qwen3.8-flash |
 deepseek-v4-pro-0813 | deepseek-v4.1-flash | glm-5.2 |
| ZHIPU/GLM-5.3 |  kimi-k3 |  MiniMax-M2.5 |
Image & video
Understanding
Extract text descriptions or structured data from images and videos
| qwen3.8-max | qwen3.7-plus | qwen3.8-omni-flash |
| kimi-k3 |
Generation
Generate images and videos from text or images, with support for editing, reference, and high-resolution output
| qwen-image-3.0-pro | wan2.7-image-pro |  happyhorse-1.1-t2v |
| happyhorse-1.1-i2v | happyhorse-1.1-r2v | happyhorse-1.0-video-edit |
| wan3.0-video |
World models
Real-time interactive open-world model — Adventure, Directing, and Acting
 happyoyster-1.0-adventure | happyoyster-1.0-directing | happyoyster-1.0-acting |
Audio & speech
Text-to-speech
For audiobook reading, voice broadcasting, virtual avatars, and more
| qwen-audio-3.0-tts-plus |
Speech recognition
Dedicated ASR and LLM-based approaches — choose based on accuracy and flexibility
| qwen-audio-3.1-asr-flash-streaming | qwen-audio-3.1-asr-flash-filetrans | qwen3.8-omni-flash-realtime |
| qwen3.8-omni-flash |
Speech-to-speech
End-to-end voice conversation without separate ASR and TTS calls
| qwen-audio-3.1-realtime-plus | qwen3.5-omni-plus | qwen3.8-omni-flash-realtime |
Omni
Integrates understanding and generation across text, images, audio, and video. For audio/video analysis and text generation, use qwen3.8-omni-flash through Chat Completions or Responses. For real-time audio/video conversations, use qwen3.8-omni-flash-realtime.
For non-real-time audio output, use qwen3.5-omni-plus.
| qwen3.8-omni-flash-realtime | qwen3.8-omni-flash |
Embeddings & reranking
Convert text or multimodal content into vectors, combined with reranking to improve retrieval accuracy
| text-embedding-v4 | qwen3.7-text-embedding | tongyi-embedding-vision-plus |
| qwen3-rerank |
Decision Model
A structured decision model for high-frequency business decisions. It performs classification, yes-or-no decisions, and scoring in a single forward pass, returning probability distributions and confidence.
| decision-model-preview |