The CMS Collaboration developed a particle-flow (PF) reconstruction algorithm tailored to the CMS detector, which has been used in physics analyses for the first time at a hadron collider. The PF approach provides a comprehensive list of final-state particles, enabling a global event description that significantly improves jet and hadronic τ decay reconstruction, missing transverse momentum determination, and electron and muon identification. This method also facilitates the identification of particles from pileup interactions and efficient pileup mitigation. Data collected at 8 TeV center-of-mass energy show excellent agreement with simulations, confirming the superior PF performance up to an average of 20 pileup interactions. The CMS detector is well-suited for PF reconstruction due to its large magnetic field, fine-grained tracker, hermetic electromagnetic calorimeter, hadron calorimeter, and excellent muon spectrometer. The detector's properties, including the magnet, silicon inner tracker, electromagnetic and hadron calorimeters, and muon detectors, are described in detail. The PF reconstruction process involves the reconstruction of charged-particle tracks, electron and muon tracks, and calorimeter clusters, with specific algorithms tailored to each component. The PF algorithm combines these elements to identify and reconstruct particles, providing a holistic view of the collision event. The performance of the PF reconstruction is compared to traditional methods, demonstrating superior efficiencies and resolutions, especially in the absence of pileup interactions. The PF approach also aids in cross-calibrating subdetectors, validating measurements, and identifying and masking detector backgrounds.The CMS Collaboration developed a particle-flow (PF) reconstruction algorithm tailored to the CMS detector, which has been used in physics analyses for the first time at a hadron collider. The PF approach provides a comprehensive list of final-state particles, enabling a global event description that significantly improves jet and hadronic τ decay reconstruction, missing transverse momentum determination, and electron and muon identification. This method also facilitates the identification of particles from pileup interactions and efficient pileup mitigation. Data collected at 8 TeV center-of-mass energy show excellent agreement with simulations, confirming the superior PF performance up to an average of 20 pileup interactions. The CMS detector is well-suited for PF reconstruction due to its large magnetic field, fine-grained tracker, hermetic electromagnetic calorimeter, hadron calorimeter, and excellent muon spectrometer. The detector's properties, including the magnet, silicon inner tracker, electromagnetic and hadron calorimeters, and muon detectors, are described in detail. The PF reconstruction process involves the reconstruction of charged-particle tracks, electron and muon tracks, and calorimeter clusters, with specific algorithms tailored to each component. The PF algorithm combines these elements to identify and reconstruct particles, providing a holistic view of the collision event. The performance of the PF reconstruction is compared to traditional methods, demonstrating superior efficiencies and resolutions, especially in the absence of pileup interactions. The PF approach also aids in cross-calibrating subdetectors, validating measurements, and identifying and masking detector backgrounds.