Media report · AI-ASSISTED BRIEF
miHoYo Unveils Gamer, a General Vision-Language-Action Model That Plays Games Autonomously
Report publication · YYSI publication
According to IT Home, miHoYo's Honkai AI research team published Gamer, a general vision-language-action foundation model that plays 2D and 3D games by simulating human keyboard and mouse input. It reads frames at 5 Hz without game APIs or internal data, was pretrained on over 10,000 hours of cross-game input traces, and uses a Streaming-vLLM engine that cut end-to-end latency from 1306 ms to 65.12 ms on NVIDIA H200. The team says it is academic research, not for cheating.
Source & context
www.ithome.com · Read the original source ↗
This AI-assisted brief attributes information to the linked media report. It is not an official announcement or independent YYSI reporting. Consult the original report and its evidence.