Cost per task is 1/20 of Hy4… Tencent launches Hunyuan Flash
Original headline과제당 비용 Hy4의 20분의 1…텐센트 혼위안 플래시 출시
Not investment advice. Muhabbit scores how much a story may matter, not which way to trade.
Tencent launched its Hunyuan Hy4 flash AI model on April 16, claiming a per-task cost one-twentieth that of the existing Hy4 in its internal workplace evaluations. The model activates 12 billion of its 224 billion total parameters during inference and supports multimodal inputs with context windows of up to 1 million tokens. API pricing is set at 0.5 yuan per million input tokens and 2 yuan per million output tokens, with cached input billed at 0.02 yuan per million tokens. The release follows Tencent's earlier teasers of both Hy4 and Hy4 flash, and the aggressive pricing matters as a direct challenge to rivals in China's cost-sensitive model API market. Watch next for third-party benchmarks of the efficiency claims and uptake among developers comparing per-task costs across competing Chinese models.
Muhabbit's AI wrote this summary in its own words from TokenPost's report and translated it from Korean. It can get details wrong, so check the source.
Named in this story
Tencent, Hunyuan Hy4 flash, Hunyuan Hy4, Marsbit