An account here; an excerpt:

Covid-19 patients who recovered from the disease still have robust immunity from the coronavirus eight months after infection, according to a new study….

[T]he new study does show that a small number of recovered people do not have long-lasting immunity. But vaccination ought to offset that problem by ensuring herd immunity in the larger population. 

The new paper studied blood samples from 185 men and women who had recovered from covid-19—most from a mild infection, although 7% were hospitalized. Each person provided at least one blood sample between six days and eight months after their initial symptoms, and 43 of the samples were taken after six months. The team that ran the investigation measured the levels of several immunological agents that work together to prevent reinfection: antibodies (which tag a pathogen for destruction by the immune system or neutralize its activity), B cells (which make antibodies), and T cells (which kill infected cells). 

The researchers found that antibodies in the body declined moderately after eight months, although levels varied wildly between individuals. But T-cell numbers declined only modestly, and B-cell numbers held steady and sometimes inexplicably grew. That means that despite decreases in free-flowing antibodies, the components that can restart antibody production and coordinate an attack against the coronavirus stick around at pretty high levels. Crotty adds that the same mechanisms that lead to immune memory after infection also form the basis for immunity after vaccination, so the same trends ought to hold for vaccinated people as well. 

And while immunity to other coronaviruses has been less than stellar, it’s worth looking at what happens in people who recovered from SARS, a close cousin of the virus that causes covid-19. A study published in August showed that T cells specific to SARS can remain in the blood for at least 17 years, bolstering hopes that covid-19 immunity could last for decades.