Discrete Particle Swarm Optimization (DPSO), Discrete Differential Evolutions (DDE), Variable Neighborhood Search (VNS) and Threshold Accepting (TA) were proposed for the no-wait flow shop problem with makespan criterion. Both DPSO and DDE adopted the natural encoding scheme based on job permutation and employed newly designed methods to produce new individual. To improve solution quality, 12 hybrid algorithms based on serial structure, embedded structure and collaborative structure were studied. Simulation results based on the well known benchmark suites in the literature showed that hybrid heuristics had better optimization performance.