AI Model Release Tracker

Tencent’s multimodal Hy4 flash awaits promised weights and independent tests

Tencent’s multimodal Hy4 flash awaits promised weights and independent tests

Hy4 flash is reported as a 12B-active-parameter multimodal model with a 1M-token context and claimed speeds above 150 tokens/second. Open weights are promised, but artifact availability and independent evaluation remain outstanding.

Sources (2)
Updated Nov 18, 2026