The layout of the distribution path of urban logistics terminals is the key to controlling transportation costs. In order to solve the path planning problem of urban low-carbon logistics, a brainstorming-adaptive large neighborhood search algorithm (BSO_ALNS) was proposed. Firstly, a low-carbon vehicle path model with capacity and time window constraints based on vehicle fuel consumption (CVRPTW) was established with the optimization goal of minimum total vehicle transportation cost. Secondly, the brainstorming algorithm (BSO) was used to improve the quality of the initial solution, and the heuristic crossover strategy was introduced to improve the quality of the global search. Using the adaptive large neighborhood search (ALNS) local search, ten kinds of damage and repair operators were designed, and the adaptive weighting mechanism was introduced, combined with the simulated annealing criterion to avoid falling into the local optimum. The performance of the BSO_ALNS algorithm was tested by selecting C, R, CR and other types of instances of different scales in Solomon. Taking the shortest path distance as the goal, the error between the BSO_ALNS algorithm solution and the historical optimal solution is within 1.5%. With the goal of minimizing the total cost of vehicle transportation, the optimal solution is obtained BSO_ALNS compared with BSO and ALNS. It is proved that the proposed algorithm can effectively solve the problem of urban low-carbon logistics path optimization.