Chinese company Tencent has released Hy4 Preview, an open-weight text input large language model featuring 770 billion total parameters and 49 billion active parameters. This release significantly expands on the previous Hy3 model from July, which had 295B total parameters and a 256,000 token context window.
- The model supports a 1 million token context window.
- It is text-only, with no vision capabilities.
- The chat template enforces two reasoning effort levels: "high" (the default) and "no_think" (reasoning disabled).
The article demonstrates the model's capabilities by generating an SVG image of a pelican riding a bicycle, noting its internal reasoning process regarding design choices like helmets and sunglasses.