An autonomous vehicle speed control method and system, a terminal, and a storage medium. The method comprises: receiving a traffic signal-based speed optimization request initiated by an autonomous vehicle; obtaining the traffic flow at each of previous n moments at the nearest traffic signal intersection on the road where the autonomous vehicle is located, inputting the traffic flow into a trained neural network model, and outputting a traffic flow prediction result of the nearest traffic signal intersection at a current moment by means of the neural network model; and obtaining road condition information of the road where the autonomous vehicle is located and the distance from the autonomous vehicle to the nearest traffic signal intersection on the road, and according to the obtained road condition information and distance and the traffic flow prediction result, calculating the optimal speed of the autonomous vehicle to pass through the nearest traffic signal intersection. The present method can provide an energy-saving speed for autonomous driving, which is of great significance for scheduling and controlling autonomous vehicles and optimizing road resources.