This paper presents two distributed protocols for achieving finite-time consensus in continuous-time multi-agent systems. The protocols ensure that the states of agents reach agreement in finite time, and they are designed to handle both fixed and switching communication topologies. The first protocol solves the finite-time weighted-average consensus problem and can be applied to systems with switching topology. The second protocol is also effective for finite-time consensus. The paper derives conditions that guarantee the protocols to solve the consensus problem and establishes upper bounds on the convergence time. It also shows that the convergence time is closely related to the communication topology, particularly the algebraic connectivity for undirected graphs. The paper uses the theory of finite-time Lyapunov stability to prove the effectiveness of the protocols. The protocols are continuous state feedbacks, but they do not satisfy the Lipschitz condition at the agreement states, which is necessary for finite-time consensus. The paper also compares the convergence rates of two systems under the same protocol but with different parameters and shows that one converges faster when agents' states differ a lot while the other converges faster when agents' states differ a little. The paper also studies the case where the topology is dynamically changing. The main results show that the proposed protocols can solve finite-time consensus problems in multi-agent systems.This paper presents two distributed protocols for achieving finite-time consensus in continuous-time multi-agent systems. The protocols ensure that the states of agents reach agreement in finite time, and they are designed to handle both fixed and switching communication topologies. The first protocol solves the finite-time weighted-average consensus problem and can be applied to systems with switching topology. The second protocol is also effective for finite-time consensus. The paper derives conditions that guarantee the protocols to solve the consensus problem and establishes upper bounds on the convergence time. It also shows that the convergence time is closely related to the communication topology, particularly the algebraic connectivity for undirected graphs. The paper uses the theory of finite-time Lyapunov stability to prove the effectiveness of the protocols. The protocols are continuous state feedbacks, but they do not satisfy the Lipschitz condition at the agreement states, which is necessary for finite-time consensus. The paper also compares the convergence rates of two systems under the same protocol but with different parameters and shows that one converges faster when agents' states differ a lot while the other converges faster when agents' states differ a little. The paper also studies the case where the topology is dynamically changing. The main results show that the proposed protocols can solve finite-time consensus problems in multi-agent systems.