
Alibaba’s open-weight image model brings editing, transparency, and multi-reference support, but commercial use needs a separate license.
Qwen3.8-Omni-Flash is positioned as Qwen’s first multimodal model for AI agents, combining audio-video processing, tool use, and sharply lower API pricing than Gemini 3.8 Flash.

Qwen3.8-Omni-Flash is described as Qwen’s first multimodal model built for AI agents. It can process audio and video together, draw conclusions, and use tools independently for tasks such as editing vlogs, translating short videos, or summarizing movies. The model’s context window spans one million tokens, giving it room to work across long or complex media inputs.

Qwen says Qwen3.8-Omni-Flash comes close to matching Gemini 3.8 Flash on audio-video tasks. The supplied benchmark image states that Qwen 3.8 Omni Flash performs on par with Gemini Flash 3.8 in multimodal benchmarks while being much more affordable. For readers evaluating models, the key takeaway is that Qwen is competing directly on both multimodal capability and cost.
Qwen3.8-Omni-Flash API pricing is listed at $0.15 per million input tokens and $0.47 per million output tokens. Qwen estimates audio input at under $0.01 per hour, while 720p video with audio at one frame per second costs about $0.20 before response costs. By comparison, Gemini 3.8 Flash is listed at $0.75 for input and $3.75 for output per million tokens at its introductory rate, with prices set to double on January 1, 2027.
The model is available through Qwen Studio, Qwen Cloud, and the API. Open-source Qwen-MM-Plugins add capabilities including video editing, speaker recognition, PDF video notes, and reusable workflows for agents such as Claude Code, Gemini CLI, and Qwen Code. Qwen-Live Harness also enables real-time interaction using a camera and microphone.

Alibaba’s open-weight image model brings editing, transparency, and multi-reference support, but commercial use needs a separate license.

Gemini reportedly reached real company systems during a flawed security test.

AI agents are moving too fast for human-only oversight, pushing labs toward AI monitors and stronger security logging.
OpenAI’s new framework aims to speed up public reporting of model misalignment cases.