Termsslug: cache-hit

Cache Hit

缓存命中

Reusing previously computed context key-value states to cut the cost of repeated prefix inference.

Curated Intelligence

1 items · Page 1/1
InfoQ 中文T2·Models
78 pts
Tencent Releases Hunyuan Hy4, a 770B Open-Source Model with Vision and Computer Use

Tencent unveiled Hunyuan Hy4, scaling its flagship LLM from under 300B to 770B parameters while adding vision understanding and Computer Use for the first time, with focused gains in instruction following, long-context comprehension and long-horizon agent tasks. Weights are slated for open release, pricing stays at RMB 6 per million input tokens and RMB 18 per million output tokens, and the model is already live across Tencent Cloud, Yuanbao, CodeBuddy and third-party platforms like OpenRouter.