SWAT model is best, because it used so many input variables such as Temperature , Rainfall, Relative humidity , Solar radation, wind speed, Soil, LULC , Slope and DEM and give results based on physical and conceptual for each sub watersheds.
The HEC-HMS model is a lump-based model, while the SWAT model is a semi-distributed model. Lump-based models consider the total basin as a “single homogeneous element".
As previously mentioned, SWAT can use more input data (topographic data, soil properties, and land use/land cover data) that are not necessarily needed in HMS.
Also, SWAT can simulate higher spatial resolutions and finer time steps (smaller than daily, which is HMS's smaller time-step I believe).
So, if you have a lot of data avalable and you seek detail, stick with SWAT. If you have a less complex, smaller scale project (and perhaps less modelling experience), then HEC-HMS may be the best option.
As it has explained in previous messages. The HEC-HMS is the lumped model and the SWAT is the semi-distributed model. The lumped model used the one value/homogeneous for the whole area while the SWAT is the physically based semi-distributed model and it assign the single discrete value to each sub-basin and the sub-basins can also be increase/decrease based on the user. There are a number of spatial raster data sets required like DEM, Soil, Landcover. Moreover, about 6 climatic variables data on daily basis is required. SWAT also generate the polygone/sub-basins and assigned the climatic station based on the thiessen polygon approach. The creation of HRU is also a very good feature in the SWAT and you can easily find the flows in/out in each sub-basin.