88Z AI, 6 October — Ant Group's Bailin team has released the new large model Ling-3.1-flash, featuring approximately 560B total parameters and 25B activated parameters, with a 1M context window. Using a hybrid architecture, the model focuses on real-world scenarios such as office work, healthcare, and software development, demonstrating the ability to handle complex tasks like building compilers from scratch.
At a glance
- Ling-3.1-flash features approximately 560B total parameters and a 1M context window, using a hybrid architecture of KDA and Gated MLA.
- The model focuses on office, healthcare, and R&D scenarios, capable of handling complex tasks such as building compilers from scratch.
The update
Ant Group's Bailin team has officially launched the new large model Ling-3.1-flash. The model has a total parameter scale of approximately 560B, with about 25B activated parameters, and supports a context window of up to 1M. In terms of technical architecture, Ling-3.1-flash adopts a hybrid design combining 7 layers of KDA with 1 layer of Gated MLA. The model focuses heavily on applications in real-world task scenarios. In the office domain, it can integrate various materials and automatically generate Excel and PPT files. In healthcare scenarios, it significantly improves information understanding and communication capabilities. In software development, it demonstrates strong execution power, having successfully built a Lua to x86-64 compiler from scratch after approximately 17 hours of continuous operation.
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.
- 百灵大模型Source published
Source dates above refer to the original announcements; the date under the headline is this article’s publication on 88Z AI.