Baalbek mayor Mustafa al-Shall confirmed strikes hit the city and its surroundings, while state media said "enemy warplanes launched a series of strikes on the Asira area of the city of Baalbek" and in a nearby town.

Earlier Wednesday, Baalbek residents rushed out of their homes after the Israeli army ordered Lebanon's biggest eastern city and its outskirts evacuated for the first time in more than a month of war.

The Israeli army urged residents of Baalbek and surrounding villages to leave immediately, warning it was preparing attacks on Hezbollah targets.

The main roads out of the city were jammed with vehicles as civilians fled in panic, an AFP correspondent reported.

Civil defence vehicles drove around the city urging everyone to leave immediately over loudspeaker. Mosques and churches delivered the same message over their loudspeakers.

The city was almost empty, the correspondent said about an hour after the evacuation warning.

Before the evacuation order, the war had forced 60 percent of its estimated 250,000 residents to flee, an official previously told AFP, while the rest were mainly crammed into the city's few Sunni-majority neighbourhoods.

"The (Israeli army) will act forcefully against Hezbollah interests within your city and villages", military spokesman Avichay Adraee said in a post on X.

The post included a map of the entire city and its outskirts.

Known as Heliopolis (City of the Sun) in ancient times, Baalbek boasts one of the world's largest complex of Roman temples -- designated a World Heritage site by UNESCO.

On Monday, Lebanon's health ministry said at least 60 people were killed in Israeli raids on the eastern Bekaa Valley, most of them in the Baalbek region.

After nearly a year of cross-border fire with Hezbollah, Israel last month ramped up strikes on the group's strongholds and then sent ground forces across the border.

The war has killed at least 1,754 people in Lebanon since September 23, according to an AFP tally of health ministry figures, though the real number is likely to be higher due to gaps in the data.