Introducing Hy4 Preview
Tencent releases Hy4, a 770B open-weight LLM with 1M token context window
Tencent launched Hy4 Preview, a 770B parameter open-weight model (49B active) with a 1M token context window — a massive jump from its July predecessor Hy3 at 295B/256K context. The model supports two reasoning modes: 'high' (default) and 'no_think'. Its release signals continued rapid scaling by Chinese labs in the open-weight space, narrowing the gap with frontier proprietary models.