Tencent releases Hy4 flash model weights for local deployment Tencent released the weights for its Hy4 flash AI model on Wednesday, two days after launching the model through its apps and cloud API, under the Tencent Hy License. Developers can download the standard model, an FP8 version and a base model for further training; Hy4 flash accepts text and image inputs, supports a context window of up to one million tokens, and has approximately 224 billion parameters with around 12 billion activated per token. The release fulfills Tencent's earlier announcement that it planned to make the model's weights available, and Tencent published deployment materials alongside the weights. Tencent released the weights for its Hy4 flash AI model on Wednesday, two days after launching it through its apps and cloud API. Developers can download the standard model, an FP8 version and a base model for further training. The release makes good on Tencent’s announcement that it planned to make the model’s weights available. Hy4 flash supports text and image inputs and a context window of up to one million tokens. It has approximately 224 billion parameters, with around 12 billion activated per token, and targets everyday productivity tasks such as research, document preparation and spreadsheet work. Tencent has published deployment materials alongside the weights under its Tencent Hy License. Tencent https://huggingface.co/tencent/Hy4-flash