88Z AI, 18 September — Alibaba Qwen releases Qwen3.8-Omni-Flash, supporting multi-modal input and 1M context, with significant performance improvements and a sharp drop in audio API pricing.
At a glance
- Performance exceeds previous gen and competitors, with audio API prices dropping over 98%
- Supports 1M context and end-to-end long-range workflows like video editing
The update
Alibaba Qwen has officially launched the new native omni-modal model Qwen3.8-Omni-Flash. This model fully supports text, image, audio, and video inputs, with the capacity to handle a context length of one million tokens. In terms of performance, its average score across thirty evaluations has improved by more than twenty-six percent compared to the previous generation, and its audio processing capabilities overall surpass those of Gemini 3.8 Flash. Meanwhile, the price for audio input via its API has dropped by over ninety-eight percent. In practical applications, Qwen3.8-Omni-Flash can end-to-end execute complex long-range workflows such as video editing, short drama translation, movie commentary, and meeting minute generation, demonstrating robust efficiency in multi-modal task processing.
Sources & editorial note
This article is an AI-assisted synthesis by 88Z AI based on the public sources below. Product claims and availability are attributed to their publishers. It does not represent independent testing or first-hand reporting.
- 千问大模型Source published
Source dates above refer to the original announcements; the date under the headline is this article’s publication on 88Z AI.