The radiation of a vacuum arc was measured in the spectral region <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$200~\text {nm} \le \lambda \le 1100$ </tex-math></inline-formula> nm. The arc was initiated in the center of the cathode by breaking the current in the auxiliary circuit and was fed by the rectangular current pulse with a duration of 10 ms. The interelectrode gap was fixed at <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$h = 4$ </tex-math></inline-formula> mm, and the butt <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$2r =30$ </tex-math></inline-formula> -mm CuCr30 electrodes were used. The arc was stabilized by an external uniform axial magnetic field (AMF). The radiation was coming out of the vacuum chamber through quartz ultraviolet (KU)-1 quartz window. The radiation detector was a silicon photodiode with a diameter of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$d =1.2$ </tex-math></inline-formula> mm. It was located on an axis intersecting the axis of symmetry of the discharge in the center of the interelectrode gap. The distance from the diode to the axis of symmetry was <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$L =1340$ </tex-math></inline-formula> mm. After amplification, the signal from the diode was recorded on an oscilloscope. The measurements were made in a range of currents from 10 to 25 kA in order to cover the modes with the anodic activity. It was found that under the conditions of these experiments, the anode activity begins to manifest itself at the end of the pulse, starting with currents exceeding 15 kA. As the current increases, the activity begins closer to the beginning of the pulse. Considering the type of the spectral sensitivity of the diode, two series of measurements were made—measurements without a filter and through a yellow filter ZhS-10 that cuts off radiation with <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\lambda \le400$ </tex-math></inline-formula> nm. The results obtained made it possible to analyze the dependence of the radiation power on the arc current. The results showed that at high current densities in the developed vacuum arc with anodic activity, a significant part of the power is transferred by radiation. At the end of a pulse with a current of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\approx 25$ </tex-math></inline-formula> kA, the radiation power is <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$P_{\mathrm {uv}}~\approx ~150$ </tex-math></inline-formula> kW, while in the discharge, <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$P_{a} ~\approx ~800$ </tex-math></inline-formula> kW is released. That is, at high current densities at the end of the pulse, the radiation power reaches almost 20% of the power released in the arc at this time. Estimates of lateral losses due to radiation are made.