Alibaba’s Qwen to Open-Source Qwen3.8-Flash-Next, Previewing Qwen4 Architecture · TechNode
7 Articles
7 Articles
Alibaba’s Qwen to open-source Qwen3.8-Flash-Next, previewing Qwen4 architecture · TechNode
Alibaba’s Qwen team is scheduled to open-source Qwen3.8-Flash-Next and an FP8 version at 11 p.m. Beijing time on Aug. 26. The model is described as a multimodal mixture-of-experts system built on the next-generation Qwen4 architecture. Qwen said the early release is intended to give the developer community an opportunity to prepare for the full Qwen4 model family. The team has not yet disclosed further details about the model’s specifications or…
Alibaba Prices Qwen3.8-Flash API At $0.16 Per Million Tokens, Cutting Inference Costs For 125B-Parameter Model
The Qwen Team has released Qwen3.8-Flash-Next, an open-weight multimodal mixture-of-experts model that serves as an early architectural preview for the upcoming Qwen4 series. The model balances substantial capacity with exceptional cost efficiency, featuring 125 billion total parameters and an additional 51 billion N-gram embedding parameters while activating only 6 billion parameters per token. The release follows the precedent established by Q…
Alibaba unveils latest Qwen model to boost global AI adoption
Alibaba's AI advancements could democratize access to powerful AI tools, fostering innovation and competition in the global tech landscape. The post Alibaba unveils latest Qwen model to boost global AI adoption appeared first on Crypto Briefing.
Coverage Details
Bias Distribution
- 50% of the sources are Center, 50% of the sources lean Right
Factuality
To view factuality data please Upgrade to Premium










