Qwen3.8-Omni-Flash
AI model · Multimodal model · Text → Text by Qwen
Omni Senses. Agentic Delivery.
It enables autonomous task planning and tool calling across text, image, audio, and video modalities for coding and knowledge work.
Model specs
The full spec card (context, pricing per token, modalities and more) on AntSpace. Sign up free to see the full spec card.
- Launched
- 2026-09-18
- Modalities
- Text, Image, Audio, Video → Text, Code, Tool Calls
- Released
- 2026-09-18
- Modality
- Omnimodal (Text, Image, Audio, Video)
- Release Date
- September 18, 2026
- Primary Focus
- Agentic Delivery and Productivity
- Native omnimodal model for text, image, audio, and video
- Advances from content understanding to task planning and tool calling
- Optimized for coding and text-based knowledge work agents
- Released on September 18, 2026
Overview
It is designed to strengthen agent capabilities in real-world productivity scenarios, moving beyond simple content understanding to active task planning and tool calling. The model can process and analyze text, images, audio, and video simultaneously to complete complex creative work. It is optimized for agentic delivery in fields such as coding and text-based knowledge work. As a foundation for intelligent agents, it enables the manipulation of various data types to achieve complex goals.
Core Capabilities
- Agentic Delivery: It advances beyond content understanding to active task planning and autonomous tool calling.
- Creative Task Completion: The model is optimized for completing creative work by integrating multiple modalities and tool use.
- Productivity Focus: It is specifically designed to strengthen agent capabilities in real-world productivity scenarios.
Productivity and Agents
- Coding Proficiency: The model builds on general agentic capabilities for software development and programming tasks.
- Knowledge Work Optimization: It is designed for complex text-based knowledge work and information processing.
- Agent Foundation: It serves as a foundation for building intelligent agents that interact with and manipulate various data types to achieve goals.
Use cases
- Developers build intelligent coding agents that can plan tasks and call tools autonomously.
- Knowledge workers use the model to process multiple data types simultaneously for complex analytical reports.
- Creative professionals integrate text, images, and audio to complete multi-modal creative projects.
Specifications
- Modalities
- Text, Image, Audio, Video → Text, Code, Tool Calls
- Released
- 2026-09-18
- Modality
- Omnimodal (Text, Image, Audio, Video)
- Release Date
- September 18, 2026
- Primary Focus
- Agentic Delivery and Productivity
Added by AntSpace from public sources (qwen.ai).