Renewable with strong storage. Essentially, what we are currently doing to improve. Besides, according to what I heard from people who actually build plants,

Renewable with strong storage. Essentially, what we are currently doing to improve. Besides, according to what I heard from people who actually build plants, it takes about 20 years from starting the process to producing electricity. If that is true, at this point it might be too late to bet on nuclear. Had we started the process of building new nuclear 10 years ago, some new reactors might have been on-line soon-ish, and they be contributing in time to hit our current targets. But if we start the process now, the reactors will be standing, at best in 2030, probably more like 2040. That is too late to hit the Paris agreement targets. Meanwhile, we see that wind and solar (in Europe) are getting ever bigger. This is meaningful progress about curbing carbon emissions, which is what we need short term. With that come rightful questions of at what percentage of power coming from wind and solar does the unpredictability, uncontrollability, and instability of those sources becomes to big of an issue. There are probably ways to get that percentage up by adding more storage, but we can probably not push it to 100% solar and wind. So 'what do we do long term' is an important question. However, it is a different and less pressing question compared to "how do we drop emissions right now, as soon as possible". For the long term, my guess is massive over-production of renewable electricity going into large, cheap but inefficient storage for both short-term and seasonal dips. I think there will be minor gains from people choosing to use their electricity when there is much excess, but only minor. Similarly, I don't think more reliable sources like gas, or nuclear will have a place. Simply because I think they will not be capital-efficient compared to dumb cheap inefficient storage. By cheap inefficient storage I am thinking about 50% efficient methods that can easily store MegaWattHours. Things like hydrogen, air-pumped tunnels, liquid metal batteries, seem interesting. When you get payed to take out electricity, and also get payed to deliver it back, you can afford a 50% round-trip loss.