GLM-4.6V just dropped. - Reasoning + vision, plus native tool calling. - Two sizes: 10B Flash and 108B. - 128K context. Transformers + vLLM ready 🤗🔥