Abstract: Super-SBM model was adopted to measure the ecological efficiency of the Yili River Valley from 2000 to 2017, analyze the valley’s spatial and temporal characteristics and evolution law of ecological efficiency from three aspects: time series, spatial evolution and spatial correlation, and by using Tobit model, verify the influencing factors and degrees of ecological efficiency of the valley. The results showed that: ① the overall ecological efficiency of the Yili River Valley was of “WV” type; the pure technical efficiency, “V” type; and the scale efficiency, “WV” type; ② The spatial patterns of ecological efficiency varied among regions over time: in 2016, Qapqal Xibe Autonomous County moved from the cold spot area to the hot spot area, and Yining City, Huocheng County, Qapqal Xibe Autonomous County, Zhaosu County and Nilek County were in the hot spot area; ③ The overall spatial correlation of eco-efficiency showed a gradually increasing trend; in Qapqal Xibe Autonomous County, Tecs County and Yining County, agglomeration tendency was relatively pronounced; and the spatial patterns mainly varied among the H-H, L-H, L-L agglomeration areas; ④ The influencing factors of eco-efficiency in the Yili River Valley had various impacts on different efficiency types: the built-up area, gas penetration rate and fixed asset investment had a significant positive impact on eco-efficiency, while the employee numbers and industrial soot emissions had a significant negative impact on eco-efficiency. By analyzing the spatial and temporal evolution characteristics of eco-efficiency and the influencing factors, we aim to promote the adjustment and optimization of industrial structure and provide empirical references for the sustainable and coordinated development of regional eco-efficiency.