This study was done at Mlughi Agricultural demonstration farm in Bura, Taita Taveta County-Kenya to test upland rice varietal performance and di-ammonium fertilizer effects on panicle characteristics and yield parameter. A randomized complete block design (RCBD) was used with seven upland rice varieties and three fertilizer levels. The treatments were replicated thrice to give a total of 63 treatment combinations or plots. The experiment was carried out for two seasons (short rains from October to December 2018 and long rains from March to May 2019). The upland rice varieties included 17KH09010014B, 17KH09010093B, AT 058, China Hybrid, NERICA 1, NERICA 4, and NERICA 10. The fertilizer treatments were di-ammonium phosphate fertilizer (DAP), farm yard manure (FYM), and a control of no fertilizer. Topdressing was done at the flowering stage with NPK fertilizer. Data collected included plant height, number of leaves, tiller number, panicle length, spikelets count per panicle, grain yield ha-1 and weight of 1000 grains. The data were subjected to multivariate analysis of variance (MANOVA) using R version 3.6.1 program and Tukey’s HSD test, p ≤ 0.05 was used to separate the significant means. The results showed that there was significant influence (p = 0.0001) of overall seasonal, treatment and variety on the yield of rice grains in tonnes per hectare. The 2018 season had the lowest tonnage compared to the 2019 season. Crops under the control treatment yielded the least tonnage per hectare while the DAP treated fields yielded the highest in both seasons. NERICA 4 was the highest yielding variety while AT 058 was the least yielding among the treatments in both seasons. It was therefore concluded that the application of DAP and FYM significantly increased rice productivity. Farmers are therefore encouraged to carry out soil fertility testing occasionally and to plant NERICA varieties.