Grid assets are usually measured by the size of their grid connection. These are lithium ion batteries, and in the 100+MW power capacity, you will usually have 2-4 hours of duration, as the economic use case for something that big usually requires replacing gas peakers to some degree.

Prior generations of lithium ion on the grid were more used for frequency regulation, and would be far smaller and have far higher power/energy rations, like 15-30 minutes. Though this was an extremely profitable market for a while, once people figured out how easy it was to get batteries to do it the market was flooded and frequency regulation does not take a massive amount of battery to accomplish.

Though lithium ion is generally viewed to cap out at 4 hours of duration, I'm thinking that it may get cheap enough per kWh of capacity to install undersized inverters and go to 8-12 hours of capacity. This could compete with other emerging battery technologies targeting that length of duration. An early test of this will be the "long duration storage" component of the replacement package of Diablo Canyon; I dont think that a particlular vendor has been chosen, but most people seem to think that it will be non-lithium-ion that will win the bid. There are other early stage battery startups with ~100 hour duration chemistries. All of these vary based on round-trip efficiency, cost per kWh of energy capacity, and lifetime over cycling.

This won't be possible in all homes, batteries are a security risk so for sure you need safe conditions and space to install similar on how you need to pass inspections for gas. The only way I could see it working is introducing an increasing electricity tarigf. First 50Kw are cheap, next are 25% more expensive, next 50% etc (this are random number I don't stand behind them)