# Qwen3.5
Latest news and articles about Qwen3.5
Total: 4 articles found
Chinese AI Chipmaker Taichu Adapts GLM‑5.0 and Qwen3.5 to Its Homegrown T100 Accelerator
Taichu (Wuxi) Electronics has completed deep adaptation of GLM‑5.0, Alibaba’s Qwen3.5 and DeepSeek‑OCR‑2 to its in‑house T100 accelerator, enabling these open models to run efficiently on domestic hardware. The work advances China’s effort to build a full AI software‑hardware stack and reduce reliance on foreign GPUs, though performance parity with global leaders remains an open question.
China’s Haiguang DCU Achieves Day‑Zero Support for Qwen3.5, Easing Enterprise LLM Deployments
Haiguang’s DCU has completed Day‑0 adaptation and deep tuning for the Qwen3.5‑397B‑A17B model, offering immediate, pre‑optimised deployment for developers and enterprises. The achievement shortens deployment times and bolsters China’s domestic AI hardware–software stack, though independent performance validation and production testing remain necessary.
Alibaba’s Qwen3.5 Claims Gemini‑3‑Pro Parity at a Fraction of the Cost — A Shift from Scale to Efficiency
Alibaba has open‑sourced Qwen3.5‑Plus, a 397B‑parameter multimodal model the company says matches Gemini 3 Pro’s performance while operating with only ~17B activated parameters and much lower inference costs. The model emphasises architectural efficiency, native multimodal pretraining and agent capabilities, and forms part of a flurry of Chinese model launches that shift competition from raw scale to systems and cost efficiency.
Alibaba’s Qwen3.5 Appears in Hugging Face Transformers — A Quiet Move That Could Amplify China’s AI Reach
A pull request adding Qwen3.5 to the Hugging Face Transformers repository surfaced on February 9, signaling Alibaba’s latest model is being integrated into the world’s primary open‑source AI toolkit. Whether the change includes usable weights or only interface support, the move lowers barriers for developers, broadens the model’s reach, and raises questions about licensing, safety and geopolitics.