The big data era requires ultrafast, low‐power and silicon‐compatible materials and devices for information storage and processing. Ferroelectric tunnel junctions (FTJs) have emerged as low‐power and fast memristors; but their operating voltages are high for fast switching speeds, and performances of silicon‐compatible FTJs are significantly poorer than those of perovskite FTJs. Here, FTJs based on SiO2/Hf0.5Zr0.5O2 composite barrier and both conducting electrodes are designed and fabricated on Si substrates. The FTJ achieves the fastest write speed of 500 ps under 5 V (∼2 orders of magnitude faster than reported silicon‐compatible FTJs) or 10 ns speed at a low voltage of 1.5 V (the lowest voltage among FTJs at similar speeds), low write current density ∼1.3 × 104 A cm−2, 8 discrete states, good retention > 105s at 85 °C and endurance > 107. In addition, it provides a large read current (∼88 A cm−2) at 0.1 V, ∼2 orders of magnitude larger than reported FTJs. Interestingly, in FTJ‐based synapses, gradually tunable conductance states (128 states) with high linearity (<1) are obtained by 10 ns pulses of <1.2 V, and a high accuracy of ∼91.8% in recognizing fashion product images is achieved by online neural network simulations. These results highlight that silicon‐compatible HfO2‐based FTJs are promising for high‐performance nonvolatile memories and electrical synapses.This article is protected by copyright. All rights reserved