Latest / Models & Agents — Video Edition
Ep 158: Tencent just dropped a 770B-parameter open-weight MoE model with a 1M-token context and…
Models & Agents Tencent just dropped a 770B-parameter open-weight MoE model with a 1M-token context and explicit reasoning controls that builders can toggle today. What You Need to Know: Tencent released Hy4 Preview, a 770B total / 49B active parameter text-only model with a 1M token context window now available on Hugging Face. The model ships with a chat template that defaults to "high" reasoning effort and supports a "no_think" mode for faster responses. Builders should test it on long-context reasoning tasks where the explicit reasoning trace and parameter scale may shift the…
The skinny
The skinny isn't ready yet — notes appear once the transcript is processed.