Intercrops have been associated with greater yields and pest and weed control compared with sole crops. In this field experiment, we investigated agronomic performance and weed suppression by three crops—spring wheat (Triticum aestivum), canola (Brassica napus), and field pea (Pisum sativum)—alone and in all possible combinations at two sites in Manitoba, Canada, from 2001 to 2003. Crop treatments were planted at the same total density (144 seeds m−2). The effects of the different crop combinations on weed recruitment and biomass and crop production were studied in both the presence and absence of in-crop herbicides. The agronomic performance of intercrop and sole crop treatments varied greatly across site-years. Some intercrop treatments (e.g., wheat–canola and wheat–canola–pea) tended to produce greater weed suppression compared with sole component crops, indicating synergism among crops within intercrops with regard to weed suppression. Intercrop treatments resulted in land-equivalent ratios (LER) > 1 (i.e., overyielding) in both the presence and absence of in-crop herbicides. In the presence of herbicides, canola–pea was the most consistent intercrop treatment in terms of overyielding for grain (mean LER = 1.22), whereas in the absence of herbicides, wheat–canola–pea produced the most consistent overyielding frequency for dry matter production (mean LER = 1.28). In the presence of herbicides, overall grain yield stability was greatest for the wheat–canola–pea intercrop treatment. Results indicate that annual intercrops can enhance both weed suppression and crop production compared with sole crops.