B.AI Rolls Out Full Free Model Lineup: Six Flagship Models Now Accessible at Zero Cost

On August 29, B.AI announced that its free model lineup is now fully available, covering six top-tier models: GLM-5.3-Flash, DeepSeek-V4-Flash, DeepSeek-V4-Flash-Vision-Exp, Hy3 (Tencent Hunyuan), MiMo-V2.5 (Xiaomi), and Qwen3.8-Flash — all open to users at no cost.

Among them, the debut model GLM-5.3-Flash, previously known as the widely discussed "Ox Alpha," is the first native multimodal model in the GLM-5 series. It features 320B total parameters with 18B activated parameters, adopts a sparse + linear attention hybrid architecture, and supports a 1M ultra-long context window, delivering rapid response times and strong reasoning capabilities, making it a cost-effective choice for high-frequency usage scenarios. The free lineup spans pure text, multimodal, code generation, and Agent orchestration scenarios, allowing both developers and heavy AI users to access cutting-edge models at zero cost via the B.AI platform — simply log in at chat.b.ai/chat to start an efficient workflow.

[TechFlow]