Recently, I have been paying attention to robotics, and I think it is a big opportunity at the track level. I happened to see some teachers in the AI field and Virtuals teachers discussing the @codecopenflow $CODEC project, and I saw its significance for RobotCIS between the texts. codec is a project that has been issued on SOL, following a model concept called VLA (vision-language-action). In fact, it is very simple to understand, that is, recognition-instruction reasoning-action. Codec encapsulates this entire logic as an operator that can run between clouds, edge devices, computer desktops, and robot hardware, and then divides the framework into machine (the underlying infra that provides compute), system (management runtime), and intelligence (the AI decision-making core that hosts the operator). Like a person, observing the environment and the goal - understanding based on observation and instructions - to act. The operating logic of the VLA model is actually very natural, observing through input devices such as screenshots, cameras, and sensors, reasoning based on observation and human instructions based on visual language models, and then executing actions through UI interaction or hardware control. Mapping to the field of robotics, it can be understood that the operator obtains information based on the robot's perception hardware, and the agent sends specific operation instructions after understanding, such as putting the red ball into the box on the desktop. As for the economic model, it mainly consists of operators and a marketplace for computing resources, publishing operators created by themselves, and contributing computing resources such as GPUs to the network to earn $codec according to usage. It seems that the project basically meets all the feasible solutions in terms of technical ideas and business models, so the concept of the project is very large. It depends on the team, it is a developer related to @Elixir_Games and Lerobot (based on PyTorch's Realworld Robotics Framework), which is very closely related. Let's follow up and take a look, the technical narrative and understanding are more in line with my expectations, and it is not used as a token investment opinion~