You need a rough complete rocket design to compare propulsion systems. If your Mars mission has a 10 km/s delta vee, and your exhaust velocity is 10 km/s (ISP

You need a rough complete rocket design to compare propulsion systems. If your Mars mission has a 10 km/s delta vee, and your exhaust velocity is 10 km/s (ISP = 1000 s), then your velocity ratio is 1 and your fueled to empty mass ratio is e^1 = 2.7. Let's round that to 3. Ie your rocket is 2 parts fuel and 1 part other stuff like engines and tank and payload. If you need initial acceleration of say 1 km/s per day, then in SI units your acceleration is 0.01 m/s^2. If you have a nuclear thermal engine with a thrust to weight of 3, that means 30 m/s^2 or 30 N per kg. For a 100 ton empty mass, you need a 300 ton initial fueled rocket. To accelerate, you need 300000 x 0.01 = 3000 newtons of thrust, meaning only a paltry 100 kg of nuclear rocket engine! If your tanks have a mass ratio of 10 (hydrogen means this might not be trivial), you have 20 t of tanks. That leaves 100-20=80 t for payload. (The engine is in the noise.) This would be excellent. With a chemical rocket, your specific impulse would be a lot worse, your propellant mass would be a lot more (3-4x), but the tanks would contain mostly liquid oxygen by weight and thus could be lighter per contained propellant mass so it might not be as bad. Engine thrust to weight might be 10x so total engine mass could still be lower, but it doesn't dominate in this kind of slow acceleration mission. With a nuclear or solar electric rocket, you might have some weird propellant - availability and tank mass depends on that. The engine is going to be really heavy. To me, a nuclear thermal makes sense for Mars missions - it is a good fit for the problem. You still need the chemical rockets for takeoff and landing at each end. Also if you assemble in LEO, there's the question of passage through the radiation belts if you accelerate really slowly. It might work for lunar work as well, haven't looked at it.