Qwen3.8-Flash-Next
Qwen 团队发布 Qwen3.8-Flash-Next 开源模型，该模型拥有 1250 亿参数但仅激活 60 亿参数。作为多模态混合专家（MoE）模型，它被视为 Qwen4 架构的早期预览版本。用户已在 DGX Spark 上测试该模型的 Unsloth 量化版本，包括 72.5GB UD-IQ1_S 和 78.9GB UD-Q2_K_XL 版本，其中 UD-Q2_K_XL 在推理任务中表现最佳。
SummaryAI generated
The Qwen team released the Qwen3.8-Flash-Next open-source model on August 26, 2026. This model adopts a multimodal hybrid expert (MoE) architecture with a total of 125 billion parameters, but only 6 billion parameters are activated, making it an early preview version of the Qwen4 architecture. Users have tested the model’s Unsloth quantized version on the DGX Spark platform, including versions with 72.5GB UD-IQ1_S and 78.9GB UD-Q2_K_XL. The test results show that the UD-Q2_K_XL performs best in inference tasks.
Related eventsRELATED EVENTS
- 2026-09-07 21:02The `GPT-6 Astra model was released, and OpenAI President Brockman confidently declared, “Welcome to the era of AGI.”
- 2026-09-08 19:21OpenAI released GPT-6 Astra, marking the arrival of the AGI era, with capabilities for autonomous computer operation and scientific research.
- 2026-09-08 22:40① Nvidia CEO Jensen Huang posted that OpenAI’s GPT-6 Astra, released last week, was trained using approximately 100,000 NV Link 72 clusters, and he believes that General Artificial Intelligence (AGI) has officially arrived; ② GPT-6 Astra can directly operate computers and software to perform complex tasks such as programming, reaching the most advanced level in multiple fields. OpenAI has announced the beginning of the AGI era.
Key entitiesKEY ENTITIES
DGX SparkQwen3.8-Flash-NextQwen4Unsloth
Event frameEVENT FRAME
Launch
Qwen Qwen3.8-Flash-Next 发布125B参数多模态MoE模型
Coverage · reports per dayLANGUAGE SPLIT
Entity relations
SignalsSIGNALS
Keyword heat